Periodic Reporting for period 2 - I-CARE4OLD (Individualized CARE for Older Persons with Complex Chronic Conditions at home and in nursing homes)
Periodo di rendicontazione: 2022-12-01 al 2024-05-31
Optimal care and treatment for older persons with complex chronic conditions (CCC) is challenging. Care for older home care (HC) recipients and nursinghome (NH) residents is further challenged by the frequent presence of impairments in physical, cognitive, social and emotional functioning. To optimise care decisions, tools are needed for better prognostication and treatment guidance.
The main aim of the I-CARE4OLD project is to develop decision support for better prognostication of (i) health trajectories and (ii) treatment impact in older care dependent persons with complex chronic conditions (CCC) receiving HC or living in NH, using real-world high-quality data. We specified our aim into five objectives:
1. Create enriched real-life high-quality datasets on older care recipients with CCC.
2. To identify homogeneous groups of persons sharing common disease patterns.
3. Develop and validate predictive models for specific health outcomes.
4. Identification of pharmacological and non-pharmacological interventions modifying health trajectories.
5. Develop and pilot test individualized prognostications.
The main aim of the I-CARE4OLD project is to develop decision support for better prognostication of (i) health trajectories and (ii) treatment impact in older care dependent persons with complex chronic conditions (CCC) receiving HC or living in NH, using real-world high-quality data. We specified our aim into five objectives:
1. Create enriched real-life high-quality datasets on older care recipients with CCC.
2. To identify homogeneous groups of persons sharing common disease patterns.
3. Develop and validate predictive models for specific health outcomes.
4. Identification of pharmacological and non-pharmacological interventions modifying health trajectories.
5. Develop and pilot test individualized prognostications.
This period WP1 coordinated 2 review meetings, rebuttals, and prepared and submitted an amendment. WP2 relates to data management protocols, ethical risk evaluations and approvals, metadata and linkage descriptions, and harmonisation procedures. Several revisions were made to earlier deliverables following advice received in the review meetings and we continued updating the project DMP. WP3 was finalized in the reporting period 1. WP4 presents the processes and procedures implemented to create the profiles for the six targeted outcomes: death, hospitalization, functional decline, cognitive decline, increased frailty and decline in health-related quality of life. The profiles address these outcomes for older adults with CCC receiving HC services or residing in NH/long term care (LTC) facilities. For NH outcomes, the data were generated from the Minimum Data Set (MDS 2.0) while HC data were provided through implementation of the interRAI HC assessment tool. 9 new prediction models were developed, and 3 existing models were selected and refined through this effort. Logistic regression analysis was used for the prediction model development. In some cases, the model would be based on all persons, in others separate analyses were created for persons with discrete characteristics (e.g. prior hospital use). Following initial profile development, machine-learning (ML) strategies for each profile were successfully applied to further refine the prediction models. The preliminary models met reasonable eta and C Stat association criterion, with the ML models further increasing the specific accuracy of the predictive models.
WP5 aims at developing and validating ML based models to predict the impact of pharmacological interventions (PIs) on health trajectories of older adults in NH and HC. The present report includes a full description of the analytical tasks in WP5. More specifically we are reporting activities related to the development and validation of ML algorithms to predict the effect of pharmacological treatments on health outcomes. A model training pipeline containing the processes of data pre-processing, dealing with missing data, forming training data, confounder selection, model training and evaluations and results of trained models are described. The case study investigating the effect of anticholinergic medications initiation on the risk of hospitalization is illustrated as an example of the analytical activities. The pipeline for model validation is also reported and results of model validation are described.
Current tasks/aim of WP6 is to develop and (internally) validate predictive algorithms that estimate the impact of non-(N)PIs on health outcomes. In the last period, we developed and externally validated the algorithms of a variety of NPIs in the HC and long-term care settings.
WP7 is dedicated to the development and testing of a platform designed to provide individual risk predictions for patients with multiple chronic conditions. This work package focuses on evaluating the usability and feasibility of the tool through pilot testing, while also exploring how the decision support system impacts the decision-making processes of healthcare professionals. Additionally, WP7 aims to create educational materials in multiple languages to facilitate understanding and effective use of the algorithms embedded within the platform.
Key objective of WP8 is to support the realization of the project’s goals, maximise its impact and extend our reach. We do this by making use of the project’s dissemination tools that have been developed in the first scientific period and according to the actions as described in the project’s Communication & Dissemination Plan. WP8 proactively seeks input from and offering guidance and support to all consortium partners to disseminate the project’s (preliminary) results to a wide variety of stakeholders.
WP9 used linked longitudinal datasets provided by CIHI, containing sociodemographic and medical or health information from (LTC) patients including older adults in Canada during the pre-COVID-19 and COVID-19 periods. This covered the WP9 projects that included (1) “Effect of COVID-19 on Functional Decline in Facility-Based Care”, (2) “Effect of COVID-19 on depressive symptoms in HC and facility-based care”, (3) “Effect of COVID-19 on quality of care and quality of life in LTC homes”, and (4) “Longitudinal analysis of the effect of antipsychotic use on behaviour disturbance in LTC homes" developed to answer, respectively, the general objectives of WP9. WP9 focused on developing robust statistical models to predict functional decline and changes in mood over time and on developing models to support decisions about the effect of pharmacological intervention on behavioural change. It is noteworthy that the statistical models developed to investigate pharmacological interventions in Canada will be validated in external databases, and this task will be carried out in the coming months.
WP5 aims at developing and validating ML based models to predict the impact of pharmacological interventions (PIs) on health trajectories of older adults in NH and HC. The present report includes a full description of the analytical tasks in WP5. More specifically we are reporting activities related to the development and validation of ML algorithms to predict the effect of pharmacological treatments on health outcomes. A model training pipeline containing the processes of data pre-processing, dealing with missing data, forming training data, confounder selection, model training and evaluations and results of trained models are described. The case study investigating the effect of anticholinergic medications initiation on the risk of hospitalization is illustrated as an example of the analytical activities. The pipeline for model validation is also reported and results of model validation are described.
Current tasks/aim of WP6 is to develop and (internally) validate predictive algorithms that estimate the impact of non-(N)PIs on health outcomes. In the last period, we developed and externally validated the algorithms of a variety of NPIs in the HC and long-term care settings.
WP7 is dedicated to the development and testing of a platform designed to provide individual risk predictions for patients with multiple chronic conditions. This work package focuses on evaluating the usability and feasibility of the tool through pilot testing, while also exploring how the decision support system impacts the decision-making processes of healthcare professionals. Additionally, WP7 aims to create educational materials in multiple languages to facilitate understanding and effective use of the algorithms embedded within the platform.
Key objective of WP8 is to support the realization of the project’s goals, maximise its impact and extend our reach. We do this by making use of the project’s dissemination tools that have been developed in the first scientific period and according to the actions as described in the project’s Communication & Dissemination Plan. WP8 proactively seeks input from and offering guidance and support to all consortium partners to disseminate the project’s (preliminary) results to a wide variety of stakeholders.
WP9 used linked longitudinal datasets provided by CIHI, containing sociodemographic and medical or health information from (LTC) patients including older adults in Canada during the pre-COVID-19 and COVID-19 periods. This covered the WP9 projects that included (1) “Effect of COVID-19 on Functional Decline in Facility-Based Care”, (2) “Effect of COVID-19 on depressive symptoms in HC and facility-based care”, (3) “Effect of COVID-19 on quality of care and quality of life in LTC homes”, and (4) “Longitudinal analysis of the effect of antipsychotic use on behaviour disturbance in LTC homes" developed to answer, respectively, the general objectives of WP9. WP9 focused on developing robust statistical models to predict functional decline and changes in mood over time and on developing models to support decisions about the effect of pharmacological intervention on behavioural change. It is noteworthy that the statistical models developed to investigate pharmacological interventions in Canada will be validated in external databases, and this task will be carried out in the coming months.
The main challenge is to improve health and healthy ageing. In many parts of the world the aged population with complex chronic conditions (CCC) grows rapidly. High quality individualised decision support for prognostications and estimating impact of treatment of persons with CCC is expected to improve patient outcomes, decrease adverse effects, and allocate sparse resources in a more rational and equal way. We specifically push the state of the art regarding the methodology to estimate treatment effects in observations cohort data by finding innovative ways to control for bias by indication.
In I-CARE4OLD we address this challenge by focussing on better evidence informed decision making for persons with CCC concentrated in home care and nursing home settings. We anticipate that our project will have direct and indirect impact on Quality of life (by optimising appropriate (non)pharmacological treatments), Quality of care (Access to high quality decision support facilitates professionals and patients to make better decisions) and cost of care (High quality prognostic information may support better informed decisions that may reduce acute admissions and postpone nursing home admissions).
In I-CARE4OLD we address this challenge by focussing on better evidence informed decision making for persons with CCC concentrated in home care and nursing home settings. We anticipate that our project will have direct and indirect impact on Quality of life (by optimising appropriate (non)pharmacological treatments), Quality of care (Access to high quality decision support facilitates professionals and patients to make better decisions) and cost of care (High quality prognostic information may support better informed decisions that may reduce acute admissions and postpone nursing home admissions).