Periodic Reporting for period 1 - 4D PICTURE (Design-based Data-Driven Decision-support Tools: Producing Improved Cancer Outcomes Through User-Centred Research)
Reporting period: 2022-10-01 to 2024-03-31
Patients with cancer often have to make complex decisions about treatment, with the options varying in risk profiles and effects on survival and quality of life. Decision-support tools have the potential to empower patients, support personalized care, improve health outcomes, and promote health equity. However, decision-support tools currently seldom consider quality of life, and their use in clinical practice remains limited.
The central aim of the 4D PICTURE project is to redesign patients’ care paths and integrate evidence-based decision-support tools. To achieve this aim, we are further developing a promising service design methodology to redesign care paths, called MetroMapping.
Box 1 describes the MetroMapping methodology and the tools we will integrate in more detail. We will better predict treatment outcomes through the innovative algorithms we are developing. And we will incorporate patient experiences, values and preferences through AI-based models. We will evaluate these decision-support tools to ensure their sustainability as well as addressing social and ethical issues. We will explore the generalizability of MetroMapping and the decision-support tools to other types of cancer and across other EU member states.
The 4D PICTURE project is carried out by a dedicated multi-disciplinary team, integrating health care research, data science, epidemiology, biostatistics, innovation and design research, health economics, implementation science, social sciences and humanities. The team is based in eight countries: Austria, Denmark, Germany, the Netherlands, Slovenia, Spain, Sweden, and the UK. The 17 partners bring together skills and knowledge to take the development and implementation of data-driven decision-support tools in cancer treatment to a higher level. In each of the eight countries, a board of patient representatives has been installed.
Box 1: MetroMapping methodology and its decision-support tools, as developed in 4D PICTURE
MetroMapping is a methodology through which patients’ care paths can be optimized and redesigned. It uses service design methodology, which aims to create a valuable “service” through a human-centred, holistic, and iterative approach acceptance. The MetroMapping methodology has different layers (see Figure 1), showing 1) all experiences with the current care path as assessed by patients, their significant others, and clinicians, 2) the entire care trajectory, 3) the information needed in every treatment phase, 4) the persons involved in care and decision-making, and 5) the physical context in which the decisions take place. Once the current, inefficient, care path is visualized, it can subsequently be redesigned to improve the health care delivery decision-making processes. Important assets are its flexibility for heterogeneous cancer care paths and its intuitively attractive visual language, enabling both engagement of patients with various levels of health literacy and multidisciplinary collaboration. (https://metromapping.org/en/(opens in new window))
Prognostic tools. A tool that helps show –with a carefully designed user-centred interface- how cancer treatments might affect survival rates and quality of life of cancer patients. It is based on robust and transparent modeling of structured, clinical data. Target group: Cancer patients, significant others, clinicians.
A conversation tool, with visual and textual components, to open-up conversations about cancer (including preferred language and personal preferences). We build on the Metaphor Menu as developed by our partners at the University of Lancaster (A ‘Metaphor Menu’ for people living with cancer – ‘Metaphor, cancer and the end of life’ project (lancs.ac.uk)).Its development is based on methods combining artificial intelligence and citizen science, using unstructured patient experience data, such as expressed in blogs and patient fora. Target group: Cancer patients, significant others, clinicians, citizens.
The central aim of the 4D PICTURE project is to redesign patients’ care paths and integrate evidence-based decision-support tools. To achieve this aim, we are further developing a promising service design methodology to redesign care paths, called MetroMapping.
Box 1 describes the MetroMapping methodology and the tools we will integrate in more detail. We will better predict treatment outcomes through the innovative algorithms we are developing. And we will incorporate patient experiences, values and preferences through AI-based models. We will evaluate these decision-support tools to ensure their sustainability as well as addressing social and ethical issues. We will explore the generalizability of MetroMapping and the decision-support tools to other types of cancer and across other EU member states.
The 4D PICTURE project is carried out by a dedicated multi-disciplinary team, integrating health care research, data science, epidemiology, biostatistics, innovation and design research, health economics, implementation science, social sciences and humanities. The team is based in eight countries: Austria, Denmark, Germany, the Netherlands, Slovenia, Spain, Sweden, and the UK. The 17 partners bring together skills and knowledge to take the development and implementation of data-driven decision-support tools in cancer treatment to a higher level. In each of the eight countries, a board of patient representatives has been installed.
Box 1: MetroMapping methodology and its decision-support tools, as developed in 4D PICTURE
MetroMapping is a methodology through which patients’ care paths can be optimized and redesigned. It uses service design methodology, which aims to create a valuable “service” through a human-centred, holistic, and iterative approach acceptance. The MetroMapping methodology has different layers (see Figure 1), showing 1) all experiences with the current care path as assessed by patients, their significant others, and clinicians, 2) the entire care trajectory, 3) the information needed in every treatment phase, 4) the persons involved in care and decision-making, and 5) the physical context in which the decisions take place. Once the current, inefficient, care path is visualized, it can subsequently be redesigned to improve the health care delivery decision-making processes. Important assets are its flexibility for heterogeneous cancer care paths and its intuitively attractive visual language, enabling both engagement of patients with various levels of health literacy and multidisciplinary collaboration. (https://metromapping.org/en/(opens in new window))
Prognostic tools. A tool that helps show –with a carefully designed user-centred interface- how cancer treatments might affect survival rates and quality of life of cancer patients. It is based on robust and transparent modeling of structured, clinical data. Target group: Cancer patients, significant others, clinicians.
A conversation tool, with visual and textual components, to open-up conversations about cancer (including preferred language and personal preferences). We build on the Metaphor Menu as developed by our partners at the University of Lancaster (A ‘Metaphor Menu’ for people living with cancer – ‘Metaphor, cancer and the end of life’ project (lancs.ac.uk)).Its development is based on methods combining artificial intelligence and citizen science, using unstructured patient experience data, such as expressed in blogs and patient fora. Target group: Cancer patients, significant others, clinicians, citizens.
In the first 18 months of the 4D PICTURE project we have accomplished the following:
We installed boards of patient representatives in each country
- We developed, translated, and pilot-tested questionnaires to be completed in each country by cancer patients and clinicians. The aim is to gain insight into experiences and attitudes regarding treatment decision making, best ways of sharing information, the possible value of decision-support tools, and positive and negative experiences with patients’ care paths. Recruitment is well under way for the patient questionnaire and is about to start for the clinician one.
- We have worked on the creation of accessible and analyzable datasets. We are performing systematic reviews to identify models that are potentially useful and external validation studies to assess their generalizability.
- We have secured access to several relevant and large accessible and analyzable corpora in English, Dutch, Danish and Spanish. We have developed algorithms towards the automatic and interpretable identification of metaphors, emotions and disease stages. Additionally, we are exploring the possible use of large language models for the classification of emotions and disease stages. Human evaluators were used to assess the effectiveness of our methods. The Spanish and Dutch design teams were trained in MetroMapping by the Dutch team. We have started with the design of the methodology for service exploration. We have coordinated with the clinical teams and selected the care trajectories to be redesigned. In Denmark, we carried out interviews, observations and workshops. These will start in the Netherlands and Spain too.
- After developing the protocol and obtaining ethical approval, we started the pre-test/post-test study to evaluate the effects of MetroMapping implementation (outcomes and costs) in each of the selected care paths in the three countries. An online patient questionnaire was built into the Castor webbased data capture system. Documents for data transfer and safety were prepared. For each cancer type, a database for the collection of data on the use of medical services and reimbursement has been conceptualized based on clinical guidelines.
- We started applying the NASSS framework for an early-stage assessment of complexities of MetroMapping and the decision-support tools through two workshops with consortium members.
- We performed a literature review on the ethics in the different 4D PICTURE work packages. Embedded Ethics activities have been carried out, including presentations at consortium meetings, lectures, a book club, and a Signal discussion group.
We installed boards of patient representatives in each country
- We developed, translated, and pilot-tested questionnaires to be completed in each country by cancer patients and clinicians. The aim is to gain insight into experiences and attitudes regarding treatment decision making, best ways of sharing information, the possible value of decision-support tools, and positive and negative experiences with patients’ care paths. Recruitment is well under way for the patient questionnaire and is about to start for the clinician one.
- We have worked on the creation of accessible and analyzable datasets. We are performing systematic reviews to identify models that are potentially useful and external validation studies to assess their generalizability.
- We have secured access to several relevant and large accessible and analyzable corpora in English, Dutch, Danish and Spanish. We have developed algorithms towards the automatic and interpretable identification of metaphors, emotions and disease stages. Additionally, we are exploring the possible use of large language models for the classification of emotions and disease stages. Human evaluators were used to assess the effectiveness of our methods. The Spanish and Dutch design teams were trained in MetroMapping by the Dutch team. We have started with the design of the methodology for service exploration. We have coordinated with the clinical teams and selected the care trajectories to be redesigned. In Denmark, we carried out interviews, observations and workshops. These will start in the Netherlands and Spain too.
- After developing the protocol and obtaining ethical approval, we started the pre-test/post-test study to evaluate the effects of MetroMapping implementation (outcomes and costs) in each of the selected care paths in the three countries. An online patient questionnaire was built into the Castor webbased data capture system. Documents for data transfer and safety were prepared. For each cancer type, a database for the collection of data on the use of medical services and reimbursement has been conceptualized based on clinical guidelines.
- We started applying the NASSS framework for an early-stage assessment of complexities of MetroMapping and the decision-support tools through two workshops with consortium members.
- We performed a literature review on the ethics in the different 4D PICTURE work packages. Embedded Ethics activities have been carried out, including presentations at consortium meetings, lectures, a book club, and a Signal discussion group.
At this stage of the 4D PICTURE project, we have not yet achieved results beyond the state of the art.