Epidemiology and Biostatistics

Leaders: Paul Burton, Juni Palmgren, Max Baur, Dorret Boomsma, George Davey-Smith, Isabel Fortier, Jennifer Harris, Cornelia van Duijn.

Aims and objectives
Biobank harmonization may be defined as: “A set of procedures that promote, both now and in the future, the effective interchange of valid information and samples between a number of studies or biobanks, accepting that there may be important differences between those studies”. Prospective harmonization is aimed at modifying study design and conduct, ahead of time, in order to render subsequent data and sample pooling more efficient and more straightforward. Retrospective harmonization is aimed at optimizing the pooling of data, samples and phenotypes that have already been collected - between studies with inevitably heterogeneous designs.

The aim of Workpackage 1 is to contribute to, and to promote, biobank harmonization across Europe from the specific perspective of epidemiology and biostatistics. In addition, working closely with other major harmonization programs, PHOEBE will share the lead in advancing such harmonization internationally. These aims will be achieved by combining a number of specific objectives and deliverables under a multi-faceted work programme involving five distinct elements (objectives).

Work programme
(1) Formal justification of the need for international biobank harmonization.
Based on a combination of a review of the status of population-based biobanking internationally and of formal sample size calculations including simulations of an evolving population-based biobank (appropriately accounting for the realistic distribution of the time to events representing disease incidence or death), we will publish two papers that will together provide a formal motivation for international biobank harmonization. These papers will be written in collaboration with scientists involved in the Public Population Project in Genomics (P3G). The papers themselves will be the deliverables for this element of the work programme.

(2) The identification, mapping out and detailed scientific description of population-based biobanks across Europe.
This will extend work already started by P3G, and by the European Science and Technology Observatory (ESTO) under the EU funded project entitled: Biobanks in Europe: prospects for harmonisation and networking. Although it should be anticipated that we will identify a small number of population-biobanking initiatives that have been overlooked previously - or are new - the predominant work under this heading will be to compile and record a comprehensive scientific description of each biobank.

In order to optimize the utility of these data for future harmonization, both within and outside the European arena, we will adopt the standardized study description questionnaire that is used by the international Observatory of P3G. The Director of the P3G Observatory is currently revising this questionnaire. The new version is being drawn up in the light of experience of data accrual to date and of the future imperatives of biobank harmonization. Scientists in PHOEBE are being consulted as part of the revision process. Pilot work already carried out at the University of Leicester indicates that with access to an appropriate study website and with the co-operation and advice of the biobank concerned, the full P3G study description questionnaire will be able to be completed by an experienced research assistant over approximately 2.5 days of full time work.

A named research assistant (Janet Jones) will be appointed to work 0.5FTE over 2.5 years in the University of Leicester. Initially she will work primarily on objective 2 for six months, but will then move on to work on prospective harmonization under objective 3 (see below). In addition she will manage/coordinate all aspects of Workpackage 1. The combined work program spanning objectives 2 and 3 (see below) will progress on the basis of email, teleconferences and annual face-to-face meetings involving relevant scientists and support staff. All study description data, in the standardized format dictated by the revised P3G questionnaire, will be transferred to, and posted on, the website of the international P3G Observatory. In order to meet the specific needs of PHOEBE, it has already been agreed that the Observatory website will be so structured that biobank searches can, if desired, be restricted to European initiatives, and that both the PHOEBE and P3G logos will be associated with the data to be collected from Europe. The key deliverable for objective 2 will be the posting on the P3G Observatory website of a completed version of the data obtained - under the format of the revised study description questionnaire - for an estimated minimum of 30 European population-based biobanks each involving 10,000 or more participants.

(3) The promotion of prospective harmonization amongst emerging biobanks.
We will formally explore the manner in which different biobanks deal with key areas of assessment (questionnaire and physical measurement) of both exposure and outcome and how these approaches may best be harmonized. This exploration will subsume both a descriptive element: “how do contemporary biobanks actually do this?”; and forward-looking inferential elements: “how ought biobanks ideally do this”? We will initially address this objective with a prospective harmonization workshop nine months into the PHOEBE project. This workshop will be organised (and funded) jointly by PHOEBE, P3G and Generation Scotland. It will bring together five biobanks that are still under design (Generation Scotland, Cartagene [Quebec], Life-Gen [Sweden], Western Australian Genome Project, INMEGEN [Mexico]) with six that already have definitive designs (UK Biobank, EPIC, Estonian Genome Project, Kora-gen Biobank, GenomEUtwin, BioHealth Norway). All of these biobanks have agreed to be involved. After the workshop, the relevant scientists, and the Leicester-based research assistant (see objective 2), will spend the remaining period of the PHOEBE project taking forward the prospective harmonization work and the development of materials in the light of the decisions taken at the workshop. Deliverables for this objective will include the workshop itself and a report about the decisions taken and the outcomes of the workshop that will be posted on the web. A final report will be prepared at the end of the PHOEBE project and will be presented at the final conference of the project and will also be posted on the web. Crucially, the final report – and, if it is deemed appropriate in the light of the outcome of the workshop, the preliminary report - will include a recommended set of common core questions and standard operating procedures (covering both phenotypes and exposures) that reflect an appropriate and valid harmonization across the particular set of biobanks that we have considered. The work undertaken under objective 3 will also form the basis of a joint publication with P3G and the participating biobanks.

(4) Statistical methods for design, analysis and harmonization of population-based biobanks.
This objective of workpackage 1 will explore the statistical methodology underpinning the design, analysis and harmonization of population-based biobanks. The work program will start with a preliminary face-to-face workshop of up to 12 international experts (to be selected both from within and outside the PHOEBE participants) who will scope out the relevant scientific questions and will decide what aspects should be addressed with high priority. A research assistant will be appointed who will attend the workshop and will then work 0.5FTE over two years (at the Karolinska Institute) to develop appropriate materials in the light of the decisions of the workshop. These decisions will be detailed in a report on the workshop that will be posted on the web. One of the specific areas of investigative focus will be the possibility of extending GENESTAT (a pre-existing web based portal representing a knowledge base for genetic statistics [www.meb.ki.se/genestat/genestat.htm]) to include methods addressed under this element of WP1. A final report will be prepared at the end of the PHOEBE project. It will be presented at the final conference of the PHOEBE project and will be posted on the web.

(5) The promotion of an active interface between scientific disciplines.
Much of the debate and discussion about the scientific role of large population-based biobanks has taken place within isolated professional groups. For example, there has been relatively limited interaction between genetic epidemiologists and traditional epidemiologists. This has led to misleading myths and misunderstandings about the true role of large biobanks, which has hampered their evolution. We intend to address this important communication shortfall in two specific ways. First, senior scientists within Workpackage 1 will continue to develop and teach the short course entitled Genetic Epidemiology for Epidemiologists that is partly based on the Lancet series with the same aims that we published in late 2005. Although this will require no specific funding under the FP6 Coordination Action, PHOEBE will consistently be designated as a sponsor of the evolving course. Other relevant epidemiological courses taught by senior scientists within PHOEBE will also be encouraged to approach the PHOEBE executive to ask to designate PHOEBE as a sponsor. Second, we have negotiated with the International Genetic Epidemiology Society (IGES) to piggyback a one-day symposium on the epidemiology of population-based biobanks to the Society’s annual congress in York in September 2007.

Deliverables
D1       Preliminary planning, strategy setting and identification of full expert groups for all
           workpackages at Initial PHOEBE Conference.
           Time: 6 months
D 2      Debriefing, presentation and discussion of final reports, and discussion of future
           strategy for all workpackages at the Concluding PHOEBE Conference.
           Time: 35 months
D 3      Final report posted on the World Wide Web.
           Time:36 months
D 4      Paper on current status of population biobanking internationally.
           Time: paper to be ready for submission at 12 months
D 5      Paper on need for harmonization of population-based biobanks internationally.
           Time: paper to be ready for submission at 12 months
D 6      Completed study description forms on 30 European population-based biobanks posted
          on P3G website.
           Time: 15 months
D 7      Prospective harmonization workshop.
           Time: 9 months
D 8      Report on prospective harmonization workshop.
           Time: 15 months
D 9      Final report on prospective harmonization.
           Time: 36 months
D 10    Paper on prospective harmonization of population-based biobanks.
           Time: paper to be ready for submission at or before 36 months
D 11    Statistical methods workshop.
           Time: 12 months
D 12    Report on statistical methods workshop.
           Time: 18 months
D 13    Final report on statistical methods.
           Time: 36 months
D 14    PHOEBE/FP6 designated as a sponsor of relevant educational courses being developed
           by senior epidemiologists within PHOEBE.
           Time: 3 months
D 15    One-day symposium on the epidemiology of population-based biobanks to be held at
           the 2007 annual meeting of the IGES in York.
           Time: 18 months

Milestones
The sole ”decision point” will occur at the Initial PHOEBE Conference when we will determine the full composition of the expert groups. Time: 6 months
M1       The discussion at, and outcome of, the prospective harmonization workshop will
           determine precisely how the work on prospective harmonization is taken forward.
           Time: 9 months
M2       The discussion at, and outcome of, the statistical methods workshop will determine
           precisely how the work on statistical methods is taken forward.
           Time: 12 months

[CREATE]

[Add new item in list]

[/CREATE]