Deliverables

Lead: Utrecht University December 2024

D2.1 Systematic review report and updates V1

D2.1 lays the first stone of the Knowledge Warehouse. Led by Utrecht University, it reports an umbrella review of the scientific evidence on patient-generated health data (PGHD) across disease areas, searching four databases and screening more than 12,000 records through the project’s crowdsourced Screenathon method to arrive at 283 papers within scope. The review also names what is missing: patient-reported experience measures, patient preference information and evidence on how PGHD improves medical device design remain thinly covered, and those gaps set the agenda for the work that follows.

Download the deliverable

Lead: PredictBy December 2024

D2.2 Practices report and updates V1

Where D2.1 looks at published evidence, D2.2 looks at what other projects are already doing. PredictBy developed an Analysis of Practice template — project overview, methodology, results, implications — and applied it to five IMI initiatives: PREFER, BEAMER, Gravitate-Health, SISAQOL and PARADIGM. The result is a first map of complementary approaches and of the places where IMPROVE can build on work already done rather than repeat it.

Download the deliverable

Lead: Utrecht University December 2025

D2.3 Systematic review report and updates V2

The second iteration completes and updates the Knowledge Warehouse. A second Screenathon screened a further 5,842 records and added 221 publications to the 266 already included, covering roughly 4,668 individual studies across 11 disease areas including oncology, cardiovascular disease, neurology and ophthalmology. The analysis is candid about what the evidence base looks like: a geographical bias towards Western countries, uneven coverage between conditions — breast cancer well documented, atrial fibrillation barely — and PROMs far better represented than PREMs or patient preference information.

Download the deliverable

Lead: PredictBy January 2026

D2.7 Practices report and updates V2

The second practices report maps five major European initiatives generating real-world evidence and PGHD — EHDEN, PaLaDIn, IDEA-FAST, ConcePTION and GREG — using the same standardised template, so that findings can be compared rather than merely collected. It draws out the transferable lessons for bringing patient-reported outcomes into value-based decision-making, and concludes that a working PGHD ecosystem needs harmonised data models, federated analytics, patient-centred design and sustained stakeholder engagement to stay credible with regulators over the long term.

Download the deliverable

Lead: Universidad Politécnica de Madrid August 2024

D3.3 Scientific, policies and practices development V1

D3.3 sets out the foundations for the trackers that keep the Knowledge Warehouse current. Led by UPM, it gives practical guidance on how the scientific, policies and practices trackers are developed, executed and maintained, and defines the structure the later versions build on. It closes by pointing forward to the technical detail and the automation that subsequent deliverables in the series take up.

Download the deliverable

Lead: Universidad Politécnica de Madrid December 2024

D3.5 Data Dashboard V1

The first version of the dashboard deliverable defines and conceptualises the IMPROVE dashboard: the core tool for putting value-based healthcare principles into practice and monitoring them across very different use cases. It sets out a structured methodology for making sure the design answers the project’s objectives, the requirements stakeholders stated, and the specific needs of each clinical setting where the dashboard will actually be used.

Download the deliverable

Lead: Universidad Politécnica de Madrid July 2025

D3.8 Data Dashboard V2

The IMPROVE project has taken another important step towards enabling value-based healthcare through digital innovation. D3.8 presents the latest progress in the design and development of the IMPROVE dashboard — a central tool for collecting, analysing and visualising patient-generated health data. Building on D3.5, this second version reflects iterative, user-centred development with the partners running the use cases in oncology, ophthalmology, cardiovascular disease, neurology and chronic inflammation.

Download the deliverable

Lead: PredictBy December 2024

D3.7 Scientific, policies and practices development V2

The second iteration of the tracker work adds technical depth and opens the question of automation. It describes how artificial intelligence and machine learning — in particular large language models — could be used to identify current scientific methods, outcomes and policies worldwide, and keep the trackers up to date without manual re-screening. It is equally clear that preliminary analysis is needed to establish the right frameworks before any model is trained.

Download the deliverable

Lead: PredictBy January 2026

D4.1 Additional data collection: needs and methodology

D4.1 sets out how historical clinical data and PGHD are identified and collected so that real-world evidence can feed the IMPROVE framework. It details the harmonisation methods — HL7 FHIR, SNOMED CT — and the governance built to satisfy GDPR and the wider European health data rules. It also describes what each data partner contributes across disease areas from multiple sclerosis to cancer and chronic conditions, and the quality assurance that has to hold before any of it supports model development or clinical use.

Download the deliverable

Lead: Tilburg University May 2025

D6.1 Report on Assessment of Stakeholder Requirements

D6.1 reports qualitative interviews with 33 stakeholders in three groups — professionals (clinicians, researchers, policymakers), patients, and technology providers. It identifies five critical stages in integrating patient-generated health data, from collection through to the decision itself, and separates the priorities everyone shares from the ones specific to a single group. Its conclusion is that no group can be served in isolation: meaningful patient engagement, support tools for clinicians and solid technical infrastructure depend on each other.

Download the deliverable

Lead: University of Applied Sciences St. Pölten December 2025

D6.2 Fit assessment between stakeholders and framework

D6.2 tests the tools IMPROVE has built against the needs recorded in D6.1, using a survey alongside a focus group. Individual tools get mixed ratings, but read as functional clusters — data infrastructure, clinical platforms, knowledge management, evidence generation — the ecosystem covers what stakeholders asked for. The gaps it names are implementation readiness and configuration support, together with better visibility and integration between work packages.

Download the deliverable

Lead: Tilburg University December 2024

D6.4 Stakeholder identification, categorization and prioritization

D6.4 is where the stakeholder work starts. Through literature review and co-design workshops it identifies seven stakeholder groups relevant to the IMPROVE platform — health care professionals, patients, researchers and policymakers among them — and sorts them by role: end users who work with the platform directly, those involved in designing and building it, and those who need to know what it produces. It then sets out the interview programme across the five disease areas that turns those groups into prioritised requirements.

Download the deliverable

Lead: Philips Medical Systems May 2026

D5.2 Use cased detailed study definition and Key Performance Indicators (KPIs)

D5.2 This deliverable describes the study designs and use-case-specific Key Performance Indicators (KPIs) that will be used to evaluate the clinical, patient-reported, and operational outcomes within each of the nine IMPROVE use cases.

Download the deliverable