D5.3 Mapping algorithms from real world outcomes to preference based measures: QoL-AD to EQ-5D
23 Oct 2020
In this report we present the development of mapping algorithms to enable the estimation of EQ-5D- 5L responses and indices from the disease specific Quality of Life in Alzheimer’s disease (QoL-AD) questionnaire. These algorithms can be used by researchers wishing to perform cost-effectiveness analyses in patient populations for whom QoL-AD data are available, but EQ-5D-5L data were not collected. The mapping algorithms are based on a multinomial regression model, and predict responses to the individual EQ-5D-5L domains first, before converting them to indices using the United Kingdom (UK) value set (response mapping). Conversion to other country specific value sets is possible. Mapping algorithms are available for both self-rated and proxy-rated data. Stata code to apply the QoL-AD to EQ-5D-5L mapping algorithms will be available to researchers on request.
D3.5 Guidelines for use of smart devices as a measure of RWE
06 Sep 2019
The implementation of digital technologies for neuropsychiatric disorders such as Alzheimer’s disease (AD) is rapidly increasing. Advances in quantitative phenotypic assessment, through digital technology, can generate potentially significant improvements in patient-centered care (for example by allowing real life monitoring of treatment efficacy). In this report, we provide an overview of digital technologies used to assess impairments occurring in AD at different stages of the disease, in particular those in the domains of cognitive abilities, social functioning, functional ability and independence, and sleep-wake patterns and night-time behaviours. Technologies reviewed include tests of cognition like executive function that can be performed on mobile devices, as well as passive functional measurements on systems such as wearables and remotely mounted sensors.
D8.4 Landscaping paper on ethics of outcome prioritisation in AD treatment
02 Dec 2019
Ethical foresight may reduce the probability of regulatory and social ‘whiplash’ by informing public and policy debates. To contribute to this process for platforms enabling medical research based on ‘real-world evidence’ (RWE), a systematic and comprehensive review of academic literature on the ethical, social, and legal implications (ELSI) of medical data repositories was conducted to identify the issues of emerging importance for this novel form of data curation and analysis. This deliverable is a first step to defining the ELSI requirements for a RWE approach to AD research. Academic literature discussing ethical and social aspects of an RWE approach and medical data sharing was systematically surveyed to produce a narrative review of relevant concepts and issues. The review aimed to address one primary question: What ethical and social issues arise in a real-world evidence approach to medical research and healthcare decision-making? In order to understand how these issues have already been identified and discussed in the context of RWE, a systematic survey of academic literature was conducted between November 2016 and March 2017. A total of 2484 non-unique sources were identified for review across the databases, with 81 sources fully reviewed. The results of the review are presented as a narrative overview, which highlights and comments upon key themes and topics in the literature. Through content analysis, seven major themes emerged from the literature: (1) informed consent; (2) autonomy and participation; (3) transparency; (4) ownership; (5) data provenance; (6) privacy; and (7) group harms and discrimination.
D4.4 Results from pilot model validation exercises
18 Dec 2018
This deliverable reports on the external validation of three disease progression models for AD dementia: Handels’ Mini-Mental State Examination (MMSE) model, Novartis’ preclinical model, and Eli Lilly’s institutionalization model. For each of these models, data sources for external validation were sought. A validation pipeline was developed to standardize the analyses across multiple data sources, reduce the workload of database custodians, and increase transparency and repeatability. The pipeline consists of the following steps: documentation of the model development and validation steps, specification of a statistical analysis plan, extraction and standardization of the data, and generation of the validation results.
D4.6 Report on results from analysis of pharmacological interventions
18 Dec 2018
The possibility of using published clinical trial data from CAMD for a comparison of the development of MMSE scores in patients with AD in cohort studies was investigated. A research design for such a comparison was developed. CAMD trial data were accessed and cleaned for the analysis and Whitehall II was identified as a cohort study containing the relevant information for a comparison. Due to delays in the access to relevant Whitehall II data, the final analysis could not be conducted.
D6.3 Guiding principles and recommendations for the development and incorporation of RWE into clinical and market access development plans for AD
04 Dec 2018
Although there are a growing number of well-reported late stage clinical trial failures in Alzheimer’s disease (AD), the introduction of a disease modifying therapy within the next 5 years may be anticipated. These treatments are likely to target AD in the earlier disease stages, unlike drugs that are currently available that treat symptoms of moderate to severe dementia. Therefore, there is a need to establish a consensus on regulatory and health technology assessment (HTA) requirements for AD as a new drug will have to go through regulatory and HTA assessments before it becomes available to patients. This paper reports the discussions and activities of the regulatory and HTA expert advisory group (EXAG) of the 2-year ROADMAP (real-world outcomes across the AD spectrum – a multimodal data access platform) project. The EXAG discussions identified a lack of consensus on validated outcomes in the earliest AD disease stages, the need for filling gaps between outcomes used across clinical trials and real-world settings, and the role real-world evidence might have in characterising the impact of a possible disease-modifying therapy on caregivers, resource use, and long-term outcomes.
D2.5 Summary of gaps between data requirements and currently available data
25 Oct 2018
In dementia research, real-world data (RWD) is increasingly being used in accordance with data collected in intervention trials, to strengthen evidence of the effectiveness of new treatments. In ROADMAP, Work Package 3 (WP3) has identified 300 European sources of RWD. WP2 have identified dementia-relevant, stakeholder priority outcomes for patients, caregivers, HTA and regulatory agencies and health professionals. Using a questionnaire, WP3 and WP4 have collated information from RWD custodians for 77 of these data sources, highlighting the outcomes that are or are not captured in their data source. This report aims to highlight gaps in the available data for each outcome. It has three main objectives, to identify: 1. which outcomes can be obtained at scale from the RWD sources identified by WP3, 2. which outcomes are challenging to obtain from any of the data sources identified in WP3, and 3. which outcomes are not adequately captured in any data source, to the best of our knowledge? Outcomes relating to cognitive abilities, functional abilities and independence, behavioural and neuropsychiatric symptoms, details of therapeutic treatment, and mortality & comorbidities are measured in many of the data sources. Fewer data sources measure significant disease related life events, medical investigations, use of health care and social services, and patient quality of life. A limited number of data sources capture information on caregiver-oriented outcomes, in particular outcomes related to the quality of the carer’s and family’s lives. A number of limitations are outlined, such as the potential for overlap of outcomes across domains, and the variability of data included in each data source
D3.6 Guidelines for combining RCT, cohort, and EHR‐based data for RWE in AD
25 Oct 2018
During the course of ROADMAP, we investigated through several Use Cases the availability, suitability and acceptability of data, methods and tools including the point of view of different stakeholders in order to use multi-modal data access platforms and create an overview of Alzheimer’s disease (AD)-relevant data which is represented by the AD Data Cube. A major aim of the ROADMAP project phase 1 was to increase understanding of the progression of AD across the disease spectrum, from preclinical stages to severe AD dementia. By integrating and combining different datasets and sources, understanding of the AD disease spectrum and its complex mechanisms can be improved.
D7.5 Final report on dissemination activities (including impact measures) and project tools developed
18 Oct 2018
This final report on dissemination activities (including impact measures) and project tools developed provides a compilation of all dissemination activities undertaken in the project, and tools produced to support these activities. The first section of this Deliverable focuses on the communication and dissemination tools. This includes an overview of the website structure and contents including the most important metrics on website usage. It also gives an overview of other relevant tools and channels such as press releases, the conference leaflet, videos as well as the project’s ResearchGate account. Further metrics include the social media account on Twitter and a detailed overview on the external newsletter’s subscriptions as well as usage. The external newsletter’s open rates are also benchmarked with information from the service provider. The second section focuses on important activities conducted during the lifespan of the project. This encompasses a comprehensive overview of presentations at conferences, an overview of delivered and upcoming scientific publications and an outline on the real-world evidence (RWE) in Alzheimer’s disease (AD) event at Alzheimer Europe’s Conference in October 2018. Furthermore, this report gives a brief summary of how the dementia community was engaged as well as involved. Lastly, it provides a brief view into the facilitation of communication outreach as well as the collaboration of ROADMAP within the Big Data for Better Outcomes (BD4BO) programme and with other Innovative Medicines Initiative (IMI) projects.
D3.3 Update of potential data sources with RWE data in Europe
18 Oct 2018
The identification and characterisation of the data sources in order to create a landscape of available real world health care data is the first step to providing an understanding of currently available data for the relevant outcomes and outlining potential gaps in currently available information about Alzheimer’s disease at various stages. Since the landscape of data should be sustainable information that is available for the current ROADMAP objectives as well as future research on the various stages of Alzheimer’s disease, it should be documented in an accessible data source catalogue with curation and search features. For that purpose, the existing EMIF AD, DPUK and EMIF EHR catalogues were selected as the preferred repositories, (https://emif-catalogue.eu, https://dementiasplatform.uk), but at the same time taking into account the different fingerprinting needs for cohort type and EHR/national register type data.
D3.4 Final report on proof of concept technical solutions for RWE data harmonisation and integration
18 Oct 2018
WP3 facilitated the work of WPs 2, 4 and 5 by identifying and providing access to relevant data sources for answering the research questions defined within ROADMAP. WP3 has developed a preliminary workflow to achieve this. After approval by the WP leads, ROADMAP researchers were required to fill in a specifically developed scientific research question form. This was sent to WP3, which then triggered the search for relevant data in all ROADMAP partner platforms. Fast access to data is possible where the data have already been uploaded to one of the platforms and data owners have given their permission.. Researchers can also identify datasets that are not yet included in one of the partner platforms. In this case, the data owner will be approached and asked for participation in the study. Eight use cases have been identified and this report shows how WP3 has made use of the existing partner infrastructures in ROADMAP to identify relevant sources. Data for the validation of the preclinical model, developed by Novartis, were identified from within EMIF AD and DPUK. EMIF AD has additionally approached cohorts that were identified as potentially interesting by WP4. Data for the validation of the model of cognitive decline in people with AD were identified by EMIF EHR, EMIF- AD and DPUK. The dementia diagnosis validation study makes use of the SIDIAP database, while a pilot study for the estimation of the costs of dementia identified six cohorts from EMIF AD. ROADMAP partners VUMC and RUG additionally worked on the feasibility of using mobile phone applications for people with AD to collect information on social communication and social exploratory measures. For the use case regarding the validation of the time to institutionalization model, two data sources were identified through the EMIF-AD Catalogue and additional literature searches. Access to these sources was obtained and analyses are currently being conducted. For the quality of life use case, suitable cohorts in the EMIF-AD and DPUK Catalogues were identified, and another data source was successfully accessed. For the Danish dementia trajectories use case, the identification of suitable EHR data sources is still ongoing. Finally, for the use case investigating transitions between different types of care, common care trajectories, and durations of different care types within trajectories of persons with dementia in the Netherlands, data from three Dutch healthcare registries were combined. These analyses are ongoing, but will be finished by the end of the ROADMAP project.
D4.3 Selection of appropriate disease models for validation
14 Sep 2018
Deliverable 4.1 “Catalogue of RWE relevant AD models and simplistic disease stage framework” previously provided an overview of published disease progression models. A systematic literature review identified a total of 40 models. For each model, contextual information (including data sources and size, disease stage, population characteristics, etc.), model outcome, and required input variables were extracted. Additionally, three unpublished models developed by the EFPIA Consortium members were reviewed and described in a similar manner. In this Deliverable, we select three models for external validation:
- Handels Kungsholmen MMSE Model (Handels RL, Xu W, Rizzuto D, et Natural progression model of cognition and physical functioning among people with mild cognitive impairment and alzheimer's disease. J Alzheimers Dis. 2013;37:357-365)
- Novartis Longitudinal Model (unpublished prevention longitudinal model describing time-to- MCI and time-to-dementia in correlation with biomarkers time course, ROADMAP Deliverable 1, 2017).
- Eli Lilly PenTAG/GERAS Institutionalisation Model (unpublished time-to-institutionalisation model, ROADMAP Deliverable 1, 2017).
COMBINED REPORT D2.3 Stakeholder generated lists of priority RWE relevant outcomes and D2.4 Disease progression and outcomes classification matrix
30 Jul 2018
The health care challenge facing Europe of an ageing population, rising costs, and more specialised treatments is nowhere more acute than for dementia, with Alzheimer’s disease (AD) as a leading cause of this neurodegenerative condition. The European Union, with the world’s most diverse and sophisticated health care systems, is uniquely positioned to develop and exploit technologies to support the collection and use of RWE. ROADMAP has brought together 26 partners from across Europe to develop a set of consensual, priority outcomes and data integration tools related to AD, as well as specific guidelines on the handling and interpretation of RWE data. WP2 has conducted a range of research, public involvement and review activities using a variety of sources and methods in order to define a list of priority outcomes that are of most relevance to key stakeholder groups, and to produce a disease progression and outcomes classification matrix that identifies priority outcomes at each stage of disease severity. A mixed methods analysis of the combined results of the WP2 workstreams has identified a core set of priority outcomes including:
- Functional ability and independence
- Patient quality of life
- Quality of the carer’s and family’s lives
- Behavioural and neuropsychiatric symptoms
- Cognitive abilities
D4.5 Report on results from ‘hypothesis-free’ disease trajectory analyses
31 May 2018
Early detection of Alzheimer’s disease is important as medication from early stages can attenuate symptoms, delay their onset, and improve quality of life. As many patients with dementia due to Alzheimer’s disease share chronic diseases and associated comorbidity patterns, we use frequent temporal disease trajectories to form more specific patient subgroups. The prior temporal disease history of these patients could possibly provide new insights for earlier detection and contribute to a core disease progression model.
D8.3 Brief on findings of ELSI focus groups for a RWE approach to AD
14 May 2018
This deliverable reports results from:
- An updated literature review on empirical research into patient and publics attitudes to the secondary use of data. (Initially was completed as part of deliverable 8.2.)
- Findings from two focus groups conducted with people with dementia and their supporters to understand patient and carer attitudes towards the ethical, legal and social implications of a real world data platform for Alzheimer’s disease.
D2.2 Report of systematic review of published and unpublished data identifying important and relevant outcomes in AD and criteria for disease progression
26 Mar 2018
Alzheimer’s disease (AD) is the commonest cause of dementia. It has an enormous global impact and cost which both continue to grow while disease modifying treatments are sought. Identifying the real world outcomes of AD that matter most to key stakeholders will help ensure that future treatments effectively improve the lives of those affected.
D3.2 First report on proof of concept technical solutions for RWE data harmonisation and integration
25 Jan 2018
WP3 facilitates the work of WP2, 4 and 5 by identifying and providing access to relevant data sources for answering the research questions defined within ROADMAP. WP3 has developed a preliminary workflow to achieve this. After approval by the WP leads, ROADMAP researchers are required to fill in a specifically developed scientific research question form. This will be send to WP3, which then triggers the search for relevant data in all ROADMAP partner platforms. If the data are already uploaded to one of the platforms, data access can be provided fast, conditional on the approval of the data owners. Researchers can also identify data that are not yet included in one of the partner platforms. In this case, the data owner will be approached and asked for participation in the study.
D6.1 Interim report on the EXAG activities and outputs (publishable summary)
04 Jan 2018
This report outlines the background and membership of the Expert Advisory Group (EXAG), which has been established under Work Package 6, to ensure the relevance and usability of ROADMAP activities and outputs to regulators and HTA agencies throughout the project.
D2.2.1 Report of SLR of data identifying outcomes in AD and criteria for disease progression
19 Dec 2017
A group of systematic literature reviews is under way to collate all available evidence on which outcomes of Alzheimer’s disease (AD) are most important, and what constitutes a meaningful delay in its progression, from the perspectives of people with AD, their carers and healthcare professionals. A series of distinct but internally consistent, sensitive searches were undertaken to capture a broad evidence base. Specific inclusion criteria were applied to refine this to a coherent, relevant subset of information which answers our research questions. ROADMAP members of WP2 have screened and are now quality appraising full-text papers in preparation for data extraction and synthesis.
D2.3.1 Stakeholder generated lists of priority real world evidence relevant outcomes for Alzheimer’s disease
14 Dec 2017
A series of stakeholder activities is in process to define a priority set of real world dementia focussed outcomes across the disease spectrum and to define what constitutes a meaningful delay in disease progression. Effective collaboration with international partners in WP2 was undertaken to design and validate three separate surveys and workshop activities. Piloting the surveys has facilitated design improvements and identified potential technical issues, enabling surveys that will increase response rates at final implementation stage. Access to numerous distribution lists and memory clinics for Europe-wide survey distribution has been successfully negotiated. Data collection has commenced in a memory clinic in Girona, European-wide data collection from electronic surveys will commence shortly.
D7.4 Interim Report on dissemination activities and communication strategy update
08 Nov 2017
The present Interim Report on dissemination activities and communication strategy update aims to provide updated information on the communication tools developed early in the project. The achievements of these tools with regard to results of dissemination activities and their impact is evaluated.
D8.1 Review of ELSI issues in RWE approach
23 Jun 2017
Ethical foresight may reduce the probability of regulatory and social ‘whiplash’ by informing public and policy debates. To contribute to this process for platforms enabling medical research based on ‘real-world evidence’ (RWE), a systematic and comprehensive review of academic literature on the ethical, social, and legal implications (ELSI) of medical data repositories was conducted to identify the issues of emerging importance for this novel form of data curation and analysis. This deliverable is a first step to defining the ELSI requirements for a RWE approach to AD research. Academic literature discussing ethical and social aspects of an RWE approach and medical data sharing was systematically surveyed to produce a narrative review of relevant concepts and issues. The review aimed to address one primary question: What ethical and social issues arise in a real-world evidence approach to medical research and healthcare decision-making?
D3.1 Overview of potential data sources with RWE data in Europe
15 May 2017
The identification and characterisation of the data sources in order to create a landscape of available real-world health care data is the first step to provide an understanding of currently available data for the relevant outcomes and outline potential gaps in currently available information about Alzheimer’s disease at various stages. Since the landscape of data should be sustainable information that is available for the current ROADMAP objectives as well as future research on the various stages of Alzheimer’s disease it should be documented in an accessible data source catalogue with curation and search features. For that purpose, the existing EMIF AD and EMIF EHR catalogues were selected as the preferred repositories, (https://emif-catalogue.eu/), but at the same time taking into account the different fingerprinting needs for cohort type and EHR/national register type data.
D4.1 Catalogue of RWE relevant AD models and simplistic disease stage framework
28 Apr 2017
This deliverable provides an inventory of existing disease progression models for mild cognitive impairment (MCI) and Alzheimer’s disease (AD) dementia. Based on a systematic literature review, a total of 40 disease progression models were identified. For each model, contextual information (including data sources and size, disease stage, population characteristics, etc.), model outcome, and input variables required by the model were extracted. Additionally, three unpublished models developed by the EFPIA Consortium members were reviewed and described in a similar manner.
D7.3 Initial set of communication tools
24 Apr 2017
The following document aims to provide an update of the different communication tools and materials that were developed during the first six months of the ROADMAP project. The initial report on communication tools and material is based on different sections. The sections include a summary of the general communication strategy and its approach as outlined in Deliverable D7.2. Initial Communication Plan, including project branding and policies. In line with the objectives stated in the initial communication plan, the developed tools, that are part of the communication framework, aim at the optimisation of the communication within the project consortium. The collaboratively designed framework includes channels and tools which are further described regarding their dissemination impact whenever possible.
D7.1 Project website
13 Apr 2017
As a key step towards increasing the visibility and outreach of the ROADMAP project to the different audiences, a project website was developed by WP7 co-leads with contributions by WP7 partners and the broader ROADMAP consortium. The aim of the website is to present the ROADMAP project as a whole and to give updates on all ROADMAP activities and publications on an ongoing basis, as part of the project’s communication and dissemination activities (WP7).
D2.1 First list of priority RWE outcomes for AD
15 Mar 2017
We have produced a first list of real-world evidence (RWE) outcomes for Alzheimer’s disease across the spectrum, having drawn upon findings from literature and following consultation with some of our partners within the Consortium who are leading experts in their fields in both academia and industry. In the next stages of our work, we will prioritize outcomes and agree on criteria for meaningful delay in disease progression through synthesizing our findings from systematic reviews, stakeholder surveys, priority setting workshops and on-going collaboration within the Consortium.