Final Report Summary - EUROCANCERCOMS (Establishing an efficient network for cancer communication in Europe.)
Executive Summary:
The lack of efficient communication among cancer health professionals, patients and policy makers remains a significant barrier to collaboration in the EU (EUROCAN PLUS Report to the European Parliament, 2008). Information overload and a very fractionated, exhaustive array of resources, networks and knowledge providers are seriously hindering the translation and application of research findings in Europe. With the constant explosion of data the cancer community can expect to face increasingly challenging times for reliable and effective scientific communication.
The aim of eurocancercoms was to establish an integrated model for a Europe-wide cancer information and policy exchange portal that would provide a functional exchange system for accurate information and intelligence, catering to the needs of health professionals, patients and policy makers. Such a model could then be subsequently applied to other areas of healthcare.
A consortium of the major European cancer community organisations in collaboration with e-communication technical experts carried out major and systematic studies of all aspects of e-cancer information space, including analysis of novel architecture, applications and tools. Surveys of the various cancer communities patients, healthcare professionals with particular foci, and novel content including specific projects on cancer clinical guidelines. The project also produced an array of major policy studies and reports into all facets of cancer communication, from open access to the state of R&D communication within academic cancer surgery (a major but orphan area).
These studies, surveys and analysis were brought together to develop and launch a novel social media platform social to EU cancer communications eCancerHub (http://www.ecancerhub.eu/) which has served as space for bringing together 4 different communities patients, healthcare professionals, scientists and policy-makers.
Project Context and Objectives:
The lack of efficient communication among cancer scientists, health professionals, patients and policy makers remains a significant barrier to collaboration in the European Union (EU). The recently completed EUROCAN PLUS project surveyed all constituents in Europe involved in some way or another with the cancer research business and concluded unanimously that information overload and an overwhelmingly fractionated, exhaustive array of resources, networks and knowledge providers are seriously hindering the translation, application and implementation of research across Europe. With the constant explosion of data we can expect to face increasingly challenging times for co-ordinated, reliable and effective scientific dissemination and communication.
The EU needs to establish an integrated model for a Europe-wide comprehensive cancer information and policy exchange portal. Such a model could subsequently be applied to other areas of healthcare, such as cardiology, neurobiology etc.
Cancer is a good model as its science is far advanced, subspecialisation within basic and clinical research has splintered the area into hundreds of boxes, and its impact on the health of Europeans is a significant issue. As the population ages, deaths due to cancer are expected to rise from around one million to almost double within a quarter of a century. Worldwide, the WHO has now declared that cancer causes more deaths globally than malaria, AIDs and tuberculosis added together. There is at present no workable model centralised or networked which even begins to address the communication challenges within this area of health science.
To address this void, the consortium will conduct an inventory of all existing information tools currently in use, including gene databases and the NCI resources in caBIG, faults and flaws and requirements for the future. E-surveys and other online research tools will be used to provide rapid and easily collated information. This will include collection of current national and international legal and regulatory requirements, for instance on Data Protection, Privacy and Ethical Issues and the provision of strategic intelligence on cancer research activities for policy makers, including those in Pharma and biotech companies. A review of new technologies available to assist in the dissemination of the information will be carried out. The consortium will seek to define and establish a state of the art communications system to connect and involve all those implicated in translation of basic cancer research into clinically testable hypotheses, public health prevention and management strategies, patient information, participation and support activities
The European Cancer Research Managers (ECRM) have been set up by the EC to conduct research of different populations in the cancer arena within Europe. They will be responsible for all e-questionnaire design and analysis, exploiting their partnership with the University College, London School of Library Archive and Information Studies SLAIS.
One of the coordinating partners, ecancer.eu will act as the medium for testing and feedback, and then will be used as the testbed for technological solutions to the problems unveiled by the consortium. Although ecancer.eu has state of the art web and TV capacity now, by the end of the project it will have to have adopted the latest technical solutions emerging from academic and industry sources. The proposed project will involve representatives from a wide range of scientific and clinical societies and disciplines (ECCO, EORTC, EACR, EUSOMA, SIOPE, OECI and ESMO). These organisations represent over 50,000 cancer scientists and clinicians from across Europe spanning molecular analysis of cancer cells to palliative care in adults and children with cancer. They will work to define the barriers, problems and requirements in publishing in a field of science which has exploded in its information content and multiplied its dissemination challenge. But the study will go further, to examine post publication information dissemination, understanding by patients. Tenovus will lead the patient (consumer) research in areas such as requirements for believable information, understandable language, and communal use and mutual sharing of support mechanisms, generated by the cascaded information in the project. Also, in the consortium are highly experienced academics and practitioners of the science and technology of communication particularly e-health, e-genetics, human computer interaction design, virtual communities, electronic villages, digital active television (Brandcast, FORTH, SIVECO and CTRC). Two global technology companies (GE Healthcare and Philips) are partners and they have immense depth in communication solutions, particularly in imaging, and each specialises in oncology.
The European Society of Medical Oncology (ESMO) will facilitate debate and provide expertise relating to communications between health professionals in oncology, consumers (cancer patients) and the pharmaceutical industry.
There are two proposed supporting co-ordinators. ECCO, the European CanCer Organisation, consists of all the principle cancer societies, from basic research to clinical care delivery, and its President Professor Lex Eggermont will coordinate the Working Parties, charged with specific Work Packages designed to inventarise scientist, clinician, pharma and consumer usage, and bottlenecks to more efficient communication. The second supporting coordinator is ecancer.eu chaired by Professor Gordon McVie. He will organise the Working Parties consisting of technical experts in web use, digital television, grid techniques, library functions, Open Access, conventional publishers and European Policy experts. Ecancer.eu is a new online, free, open access cancer journal from the European Institute of Oncology (IEO), which already receives over 12000 visitors per month. All documents arising from the consortium will be published immediately on a dedicated part of the journal, and e-questionnaires will be organised through its website.
A Scientific Committee (SC), chaired by Professor Veronesi from IEO will coordinate the whole consortium, ensuring that bridging between demand and solution is efficient and speedy. The SC will analyse the reports generated to identify common resources to optimise the best way forward, to determine priorities for fast and efficient communication within cancer research in Europe, and to integrate the ideas into a common strategy and to supervise feasibility trials on ecancer.eu.
The Coordinating Action (CA) will make recommendations for the publishing and dissemination of cancer research in the EU including identifying further priorities for the 7th and even 8th Framework Programmes. These will then be widely circulated throughout Europe prior to finalisation by the SC and submission to the European Commission. Final reports will be published in ecancer.eu and conventional journals, such as Lancet, European Journal of Oncology and Annals of Oncology. Their publishers have declined to be partners, but remain willing to participate in workgroups and dissemination activities. The CA will, among others, address the following barriers to efficient information flow:
Basic research
The need for a scientist to be original and to publish their research first. A scientists career progresses according to the number and originality of papers they publish, they are therefore unwilling to share initial findings and remain secretive about ongoing data. Negative results are infrequently published in scientific journals [1].
Cancer networks will share information, though not often raw data, within their own network; however they rarely link with other cancer networks on a functional basis. New initiatives to publish raw data directly from scientists e-lab books will be studied
3. All new discoveries require a patent to secure the intellectual property. Usually a bio-tech company is founded to prove and exploit the patent. Once commercial interests are involved the release of information becomes timed for commercial advantage. Is this a significant barrier?
4. Pharmaceutical companies carry out a large proportion of research in Europe (around 50% in the cancer area), but 80% of their science findings are never published. Commercial interests, including stock market listings influence the publishing, and the timing of publication, of scientific research. Several major Pharma companies have indicated a strong desire to be involved in developing solutions to this issue, though not as full partners.
5. Language barriers among scientists severely hinder rapid science dissemination, as the vast majority of science publications are in English and the majority of European scientists do not have English as a first language.
Technical language barriers within subsets of the oncology community create communication barriers. Patients are also excluded from understanding new research by the use of unfamiliar technical terminology.
Differing regulations and procedures in EU member states prevent cancer research being disseminated across the EU. Procedures need to be standardised to allow for the free flow of research and this will require European coordination.
Peer Review is fast approaching its sell-by-date. What is to replace it?
Clinical Research
Clinical research has the added restriction of maintaining patient confidentiality. The issues of data protection and privacy are of prime importance to the consortium. Fortunately we have two partners from another EC programme ACGT, who have already far-advanced proposals to solve these issues. The Independent Ethical Committee will take a pro-active role in policing this topic.
Conflicts of interest are of potential importance when clinical trials report results on new molecules and devices from the pharmaceutical sector.
There is ample evidence which highlights the heterogeneity of outcomes of cancer care throughout Europe. What is clear from all of them is the detrimental impact of lower social class, deprivation, ethnicity, low educational achievement, and access to relevant information amongst cancer patients. These issues are seminal to this project and the communication needs arising from them will be investigated and analysed. A variety of appropriate solutions will be developed to meet the identified needs.
Patients are increasingly using the Internet in a hunt for reliable cancer information. One website on cancer opens each day of the year, and the word cancer is searched on Google more than any other except pornography! Confusion is rife, total failure to translate published results of research is all too common, and public understanding of the reality of cancer prevention messages and access to best care (including access to clinical trials) is abysmal. Needs of patients for particularly relevant information, its optimal timing and mode of transmission will be determined.
A particular area of missed opportunity is the communication of information to patients and the general public on risk. Cancer is a genetic disease, and lack of understanding of even the most important familial genes in breast, colon and ovarian cancer has led to failure of application of important scientific discovery. Two partners have particular expertise in this difficult territory (EACR and Tenovus), and again the ethical issues are numerous and must be solved.
An important aim for the Commission is to stimulate analysis and debate to improve the standards of the European science system and its ability to produce robust and reliable knowledge to support policy-making and respond to social needs.
Project Results:
COMMUNITY ANALYSIS
Paediatric oncology
The State of Research into Children with Cancer across Europe New Policies for a New Decade was published by the European Society for Paediatric Oncology (SIOPE) and ECRM Foundation. The report into the e-communication needs of the paediatric oncology community was presented at a special event in the European Parliament (wednesday 9 February) to mark International Childhood Cancer Awareness Day to a wide range of EU policymakers as well as young cancer patients and parents.
They found that countries with a larger oncology burden, such as those in Eastern Europe, tended not to collaborate in research with those with a better-developed research structure, and this in turn affected the care they were able to give young patients. Paediatric oncologists from ten countries were asked about their national situations and the differences were striking, said Professor Kathy Pritchard-Jones, Cancer Programme Director at the UCL Institute of Child Health, London, UK and Past President of SIOPE. For example, a respondent from Bosnia-Herzegovina cited lack of hospital space, no specialist paediatric oncology radiologists, and inadequate diagnostic facilities, with everything adapted to adult patients. The Czech Republic said that research grants were usually for only three years and there was little support for young scientists to attend international courses and congresses which could help harmonise the level of knowledge across European states. In Italy, it was the fragmentation of childhood cancer care which concerned doctors, with nearly 50 centres specialising in paediatric haematology and oncology, and a lack of co-ordination between research laboratories and clinics. In comparison, Sweden has only six paediatric oncology units, which work closely together but, even there, there are problems of funding and a lack of experienced staff. The UK expressed concern about the effects of health service cuts on childrens cancer care, and the fact that infrastructure changes had moved paediatrics alongside the adult cancer model, meaning that research for children now has to compete for funds at a much more visible level with research for adults.The report also found large differences in the provision of information on childhood cancer, with variations in the involvement of parental organisations, the use of digital media, and the adoption of a common national standard for information provision. When a child is sick, the provision of accurate information to the patient and family is essential said Professor Sullivan. We believe that the establishment of a European Common Information Portal could do much to tackle major deficiencies in information in countries with few or no patient organisations, or where the existence of a large number of languages make access to such information difficult. The report called for adequate long-term EU funding to support a Europe-wide clinical trials network for paediatric oncology. This is an essential prerequisite for the effective testing and dissemination of new therapies and techniques, said Professor Pritchard-Jones. We also need to study treatment outcomes, and for this the creation of a European Childhood Cancer Epidemiological Registry is essential. With these tools we can maintain the enormous progress that has been made in the past. Without them, we run the risk of jeopardising that progress and failing children who are dependent on us for their survival.
[*Sullivan R, Pritchard-Jones K, Camporesi S, Lewison G, Vassal G, Ladenstein R, et al. (2011) The state of research into children with cancer across Europe: new policies for a new decade ecancer Vol 5 doi: 10.3332/ecancer.2011.210]
Cancer Surgery
Eurocancercoms made extensive use of bibliometrics to map out the various cancer communities operating in R&D. Here it analysed global cancer surgery R&D trends over the last ten years across 21 countries.Surgery is the main modality for cancer cure and control globally. Yet, in comparison to other areas such as cancer drugs, we know little about ongoing research activities to inform policymakers. Two sub-field filters, surgery research and oncology, were developed and applied to Web of Science. The intersection of these two filters identified papers in surgical oncology, and their bibliographic details were downloaded for analysis. This included matching of five-year citation counts to the papers, impact factor, geographical analysis by country, translational collaboration, involvement in clinical trials, citation on clinical guidelines, and percentage of reviews. Surgical oncology represents about 9% of all cancer research low in comparison with surgerys contribution to cancer treatment. The USA published the most, followed by Japan which had a high relative commitment to surgery within cancer research, followed by the large West European countries. Although Swedens papers were relatively basic it participated the most in clinical trials. Its papers were also the most cited on clinical guidelines, but contained relatively few reviews, where the UK, Greece and Belgium scored best. Surgical oncology papers are generally not well cited compared with cancer research overall, but on this measure the Netherlands, the USA and Sweden scored best. International collaboration was measured relative to what might have been expected, on this indicator Canada, Switzerland and the USA were the best performers. The results generated enabled a) new policy intelligence to be created and b) provided the spine for e-contact and networking between the active academic cancer surgery community through ecancerHub
Patients & Risk
Risk in cancercommunication: The main findings are presented as the key themes and issues arising from the RTD. There was consistency in the group around the need for improvements to the risk communication process, particularly around information and its onward diffusion i.e. Telling the family and differences regarding wanting to know their genetic status. This was evident from the responses and the more extensive discussion on these topics. The women shared their perspectives and experiences of breast cancer or risk communication and those of others in similar situations from their personal or advocacy group connections. The additional Health Care Professional perspective contributed examples of research and their knowledge of services relevant to the discussion. The richness of the data emerges from the group dynamic and their diversity as not everyone had the same views and experiences due possibly to differences in, for example age, education, and in particular in this situation, across countries.
The following themes were identified from the RTD. Key quotations are included in the text to illustrate the theme and to show that within the topic there exists a range of comments, sometimes opposing.
The participants are identified as HCP1, HCP2 (health care professionals), P3-6 (patient representatives) where P4 added comments pre or post interview = 4a and 4b, R7, R8 (Macmillan Cancer Voice reviewers)
The Round Table Discussion was introduced with a brief overview of cancer genetic services so this not only set the scene but provoked some discussion about current service provision and recommendations for potential improvements. A person entering the cancer genetic service was described as there for life due to the changing information and support needs at different stages through the journey. HCP2
There were several positive comments about the service, being non directive and allowing sufficient time to discuss things, but disappointment with lack of support when passing on the news to family members:
Twelve years ago you just got a letter with your diagnosis and to tell your family about it and Ive heard from other people its still the same P3
Involving patients in service redesign was advocated including involvement in guidelines.
How is the whole process of communication experienced by people who have a predisposition, what do they need, what have they missed and what solutions are proposed P4a
persuading, ...professionals not just to listen to women but actually to convene groups of people and to try and be informed.... looking at how populations get information and how they interpret general information...if that were pursued with more vigor certainly in Ireland I think wed have better outcomes. P5
There was recognition that a key beneficial difference with specific cancer genetic services compared to consultations with general practitioners (GP) was the planned time for discussion.
not like your six minutes with your GP in the UK HCP2
Media
From the 240 media contacts contacted to complete the survey, there were 63 responses (26.25% response rate). The respondents were first asked to give their occupation:
The majority of respondents were from the United Kingdom (42.37%), followed by the United States (18.64%). The remaining respondents came from Australia, Belgium, Finland, France, Germany, India, Italy, The Netherlands, Spain, Canada and Asia/Pacific.
The professional experience of respondents weighted heavily towards those with more rather than less experience: 53% had 20+ years of news/media experience, 12% 15-20 years, 18% 10-15 years and 17% had less than 10 years experience.
Respondents were asked in what type of news organisation they worked, with multi answers being allowed. There was representation across the all categories given:
Respondents were then asked to select their areas of focus. The large majority (71%) selected healthcare, followed by science stories related to topical news and events (47%), 31% said specific scientific subjects, 19% general science stories and 10% science events. The remaining areas of economy, general news, industry and finance accounted for less than 11%.
Respondents were asked to rank what influenced their selection of science topics. 50% ranked public interest/opinion as the highest influencing factor, followed by topicality. On the other end of the scale, industry involvement (32%) was ranked as least influential, followed by economy involved topics and organisational press releases (both 21%).
The survey also asked respondents to indicate how interested their readers/viewers are in a selection of topics. Topics around patient involvement and stories specific to cancer type are of most interest (70% and 74% respectively). Readers are also more interested in topics such as drug development, genetics, prevention and new technology than not. Local stories, however, were said to be of very little interest (49%). Respondents were then asked to state how interested in their editors/managers are in the same topics. These responses almost evenly matched those from the previous question, with patient involvement, stories specific to cancer type, and drug development being just as interesting to the Editors.
Respondents were given a list of potential problems that they may come across when reporting science news, and were asked to state how often they come across each. The following five problems were the most often/regularly encountered (in rank order):
1. Research findings overstated by media
2. Research findings overstated by PR people
3. Uncertain or ambiguous results
4. Research findings overstated by researchers
5. Technical jargon makes the news hard to find.
They were then asked how much they agree with a set of statements relating to experiences they face as reporters. The following three statements gave the strongest responses:
The fact that 65% of respondents agree/strongly agree to the idea You would appreciate a forum where you could access a pool of experts willing to comment/give quotes for your story validates the 60% of strongly agree/agree responses to the statement You long for one central source of trustworthy information. Respondents want a pool of sources rather than just one.
This is also supported by the responses when respondents were asked how they would prefer to receive potential news stories. Email was by far the most preferred means with 95% of respondents stating it as their preference. However, other sources of news delivery received a good spread of preference:
These results would suggest that if research organisations and scientists want their news to receive the highest impact in the media, then the stories have to be distributed to the whole spectrum of news outlets.
From a list, respondents were asked to tick all statements that they identified as being frustrations. In rank order, they were:
1. Press releases/sources are inconsistent in quality and reliability
2. Cited sources/PR people respond to slowly to your queries and you miss the opportunity to publish
3. Hard to distinguish really promising news among all the dross you receive
4. Scientists/researchers/PR people dont understand what interests the public
5. Scientists/researchers/PR people dont understand your story interests
6. Press releases or other stories are too complicated or technical
CONTENT ANALYSIS
Integrating cancer clinical guidelines: understanding current information provision
In total 30 European organisations in oncology were included in the survey. Of these, 21*/30 responded to the questionnaire, the majority of which represent specific medical specialties in oncology (38.1%) or disease specific organisations in oncology (23.8%). Other participating entities included an oncology pharmacy organisation, a general cancer society, an educational organisation, a nursing and a patient society. The majority of survey participants were involved in guideline development (62%), the percentage of which rose to 75% when considering the medical oncology specialty and disease specific societies alone.
The results of the survey revealed a number of serious pitfalls in the ways in which European societies are developing guidelines:
Notably, 31 % of organisations surveyed stated that they do not have guidelines or a structured process for guideline development. This also shows a lack of homogeneity in the methodologies to develop clinical guidelines. In addition, for the majority of organisations (69%), there is no methodological training for members of guideline development groups meaning that guideline developers are not specifically trained to do so.
Searches via electronic databases are the preferred method for the collection of evidence. For almost half of the guidelines developed original articles are confirmed manually. For the majority of the guidelines (85%) systematic reviews are the preferred method of analysing evidence on which a guideline is based. Experienced based evidence accounts for 69 %.
Several methods are employed in formulating guidelines. Both informal and formal expert consensus are equally adopted in approximately half of the guidelines development. An evidence-linked system weighing the evidence according to rating criteria is used in 38% of cases and a subjective review in almost a quarter of the guidelines.
The majority of guidelines (70%) are reviewed both internally and externally, compared with guidelines developed by other groups (61.5%), and endorsed by formal authorisation by a professional organisation of the target users (46.2%) the remaining guidelines are approved by the respective Societys Board. Also noteworthy is the fact that more than half of the guidelines are not piloted among end users (54%).
Importantly, the survey reveals that 62% of the guidelines do not have an expiry date. Although most of the organisations are regularly updating their guidelines, there is significant heterogeneity in the frequency of these updates which can vary from every 1, 2 or 3 years or when important scientific information becomes available. This shows that there is no systematic approach towards updating guidelines.
Most European Societies surveyed (77%), are systematically monitoring changes in the literature related to their guidelines. The regular quality systems used for the guideline program are developing and publishing criteria for good guideline development ('guidelines for guidelines') (38.6%); revision of the guidelines based on comments from the professional community (30.8%); appraisal of existing guidelines (30.8%); submission of the guidelines to a guideline clearinghouse; publication of the guidelines in a European journal and review and updating of the guidelines by an expert panel.
Another important outcome of the survey is that patients are often, in more than one third of the cases (38%), not involved in the development of guidelines. When they are involved, it is almost equally either through surveys of patients views and preferences; or reviews by representatives of patient organisations; or by participation in the guideline development group. The patients that are involved in guideline development groups are mostly selected by the relevant patient organisations (46.2%); to a lesser degree in existing patient panels (15.4%) and in a few cases through clinical centres (5,4%). None of the European societies surveyed use open advertised calls.
The majority of the guidelines (85%) developed are treatment and disease-management related. Half of them are related to diagnosis while only a minority (15%) are related to prevention or screening. In addition to ensuring appropriate clinical care which is the purpose of three quarters of the guidelines, for almost a quarter of them, the aim is also to seek cost containment. More than half of the guidelines are targeting policy makers and not solely health professionals who belong to the institutions that are developing the guidelines.
Recommendations/conclusions ..
Guidelines are essential tools for oncology professionals and patients alike. By following up to date guidelines, prevention (primary, secondary and tertiary), diagnostic procedures in cancer treatment and care can be carried out in the best possible way. As the survey demonstrates, guidelines are developed by many European cancer organisations. Since such development is both costly and time consuming for societies and their members, a more coordinated approach should be encouraged to provide up-to-date European cancer guidelines in the most cost effective and timely means possible.
Having provided an inventory of European groups developing guidelines and assessed the corresponding methodologies and approaches to guideline development based on literature and the outcomes of the survey, the following observations and recommendations can be made:
From the very outset the entity/entities developing guidelines should ensure an optimal environment for guideline development, systematically describe the various steps involved and provide a structured budget for all aspects including: methodological training of personnel involved in guideline development and subsequent evaluation; open access to specific data relating to the guidelines (full text publications); a detailed implementation plan; rigorous evaluation system and tool/s; quality assurance plan by which the impact of guidelines in clinical practice can be evaluated, as well as an easily accessible and interactive communications tool connecting all members involved in the process.
Importantly, both the focus and the stakeholders for which the guidelines are being developed must be clearly defined with greater emphasis on patient access and involvement in particular, and upon development/ revision(s) guidelines should be easily accessible for end users through a range of electronic tools and channels.
A workshop through which to formulate and establish best practice in guideline development would be favourable in order to improve guideline development across Europe. Further, since cancer related guidelines are becoming increasingly inter-and multidisciplinary, ECCO could play a key role in coordinating these efforts as the only multidisciplinary and multiprofessional cancer organisation that connects and responds to all stakeholders in oncology in Europe.
What information should ecancerHub integrate and how should quality be maintained?
An online survey of the cancer science community in Europe was conducted to discover their experiences of and attitudes to the issues surrounding academic publishing and Open Access. The responses from the two major surveys showed very positive attitudes to the Open Access publishing route; perhaps the most challenging statistic from the EACR survey is that 88% of respondents believe that publicly funded research should be made available to be read and used without access barriers.
Cancer practice guidelines (CPG) are an important instrument that translates knowledge and data into recommendations and unlike systematic reviews, commentaries and editorials, they provide conclusive recommendations to be used in action. The development of such guidelines is a scientifically rigorous task that requires subjective judgments. However, since these judgments are made by humans, they leave room for error and bias. Within the eurocancercoms project the issue of bias in CPG integration was examined by in depth policy analysis.
The sources of bias for the authors of CPGs vary and according to an article written by Allan S. Detsky, they can originate from the participants previously established experience or personal practice, which can dominate their views of evidence. Likewise, they can also originate from a possible long service to government committees or private insurance companies that could designate a different agenda on the experts minds, like the limited cost or other perspectives.
Nevertheless, the most commonly discussed form of bias and conflict of interest is financial. Such bias is usually noticed when there are ties between pharmaceutical companies and authors of CPGs. There are at least two common ways through which industry relationships can affect the content of CPGs. The first one is directly through a financial relationship between the guideline committee members and the representatives of the industry.
A recent study, aiming to quantify the extent and nature of interactions between authors of CPGs and pharmaceutical industry reported in the Journal of the American Medical Association, revealed that the 87% of the authors of CPGs had some form of interaction with pharmaceutical industry. The study highlights the need for appropriate disclosure of the financial conflicts of interest of the authors.
Another study aiming to examine the degree and type of financial ties between the authors of CPGs and the pharmaceutical industry, has concluded that ninety percent of the authors of 3 major CPGs in psychiatry had financial associations to drug manufacture companies, the products of which were explicitly or implicitly detected in the guidelines as recommended therapies. In addition, none of the financial relationships of the authors were disclosed in the CPG.
Choudhry and colleagues have documented that authors of CPGs published in the 1990s had many financial relations with pharmaceutical companies, the products of which were being discussed in the guidelines. A survey of those authors, reported that only the 7% of these authors believed that those relations influenced their opinions when the 19% of them believed that their coauthors were actually influenced.
Finally, a review, the purpose of which was to explore the extend and scope of American Academy of Neurology (AAN) guideline author, reported conflict of interests and implications for management. The report states that 46 of the 50 CPGs which were examined had at least one author with a conflict of interest. More precisely, the authors of the Quality Standards Subcommittee guidelines were more likely to have personal income conflict of interests with pharmaceutical and medical device companies whereas authors of Therapeutics and Technology Assessment guidelines were more likely to have clinical practicerelated COIs.
The second way, by which pharmaceutical industry influences the authors of clinical practice recommendations, is indirectly and lies in the process by which raw data becomes medical knowledge. Since the development of CPGs relies heavily on published evidence, there is a big chance that any industry involvement in the conduct of clinical trials and medical research publications in respected medical journals, can exercise significant influence on the authors perspective of view.
Recent literature discussing the influence of industry on medicine, reports that academic institutes are not left unaffected by the same influence, even though they are frequently regarded as free speech bastions of learning. That is also the conclusion of two former principal editors of the New England Journal of Medicine (NEJM), who state that it is almost impossible to find an expert who is not receiving payments from one or more drug companies in the filed.
Further concerns have also been expressed that pharmaceutical companies are involved in the field of research since they are capable of influencing study results. In addition there is evidence data showing that resulting outcomes from drug studies, which have been funded by pharmaceutical industries, are favorable to the funders.
Finally the integrity of medical reports seems to raise concerns as well. A recent study has revealed that 96% of the authors who wrote articles certifying the value of specific cardiac medications had a financial tie with the manufacturer. Further reports state that an amount of scientific articles appearing in widely known medical journals under the authorship of scientists and doctors, were actually written by ghostwriters on the payroll of industry.
Bibliography, reports several common forms (types) of financial influence that have been noticed in the process of developing CPGs. Pharmaceutical industry Sponsor, or other funding agencies can exercise influence in the selection or the guidelines topic, the selection of the expert panel members, the selection of technical consultants, the writing of the guidelines or even the final guideline document.
The role of social media in cancer communication
The relevance of social media to the eCancer domain results primarily from the fact that they can change (meaningfully) the way in which networking in the field is shaped, organized and maintained. A pre-requisite is to design tools, which will make users realize such benefits and internalize the new means for cooperation. Today this is a prominent challenge for professional communities, as it deviates from the way in which traditional practices are enacted and performed, while requiring a new culture for collaboration. The increasingly social conduct of work and the spin-off effects it bares on other relevant communities (i.e. neighboring disciplines, end users, etc) is clearly a motive. Nevertheless, appropriating the benefit is not straightforward.
For the cancer community the challenge is not setting yet another social networking site. Instead, it should aim to challenge the more demanding issue of crossing boundaries (or fields of expertise) and making new grounds for turning experience to collective know-how and know-why. Social media, designed suitably to foster cross-functional virtual communities and their practices in selected domains of eCancer, provide the means to this end. For the purposes of the present discussion and following the recent meeting in Crete, candidate areas for experimenting with boundary-spanning virtual practicing include:
Open access publication management
Cancer research management trials
Building evidence-based guidelines
Understanding the patients view and experience of care
Personal health records
Spanning boundaries Consider a professional research group specializing in cancer research. The group has just initiated a trial using a new drug, which seemed to work in laboratory settings. Therefore, they now wish to do the following: (a) disseminate their laboratory results to the wider community and (b) register their new trial.
Open access publication Attaining objective 1, the materials to be shared include a document, figures and images (linked to the document), sample visualizations (extracts of what was found in the lab), video, etc. The group leader enters ECCs and realizes that the available mechanisms for disseminating work include: (a) a paper following a specific template (i.e. keywords, abstract, single/double column, etc); (b) one-page leaflet following again a specified template; and (c) TV broadcasts. Choosing the preferred dissemination medium allows the group leader to populate an object (by uploading material, inserting figures, tagging, etc) and assigning some distinct services that the group would like readers to use so as to provide feedback. At the end of the day the groups work is updated and made available through a social networking site which compiles the collective wisdom on cancer research trials.
As shown, the Figure is not aimed to present merely the published material as accepted and uploaded at one point time. Instead, it emphasizes a living collection of relevant objects all contributing to gaining insight to how the initial contribution evolves with the input of third parties. As an example this presents this evolution in the fifth revision of the initial object collection. The activated links indicate the evidence added in this firth revision and how such evidence is classified (i.e. videos and pdf files).
It is important to note that the specific layout is only indicative, while we have intentionally chosen to organize the material in the manner depicted in web analysis. This organization reflects a classification of contributions by type of object (i.e. video, pdf file, figures) and keyword (i.e. drug) so as to enable further exploration within these categories. Thus, it is possible to query the available data by simply selecting interaction nodes scattered throughout the visualization. Finally, our choice of layout strategy is such that it allows us to establish links with the next scenario (see section 0), spanning the boundary between open access publication and cancer research trial management.
Managing cancer research trials Attaining objective 2, the information to be shared includes the type of trial, drug involved, treatment as a series of interventions, trial status / phase, sponsor, etc. Again, the group leader goes through some forms to do the task or in fact he can instruct the system to upload this information from another source (following a suitable schema).
The distinct feature of the visualization is that it classifies all collected experience in a manner suitable for quick overviews as well as visual exploration. Quick overviews allow users to gain insights to various data classification schemes. By presenting information on drugs (lower part) being investigated by the available cancer research trials, per phase (middle part presented as timeline) and the patient population split in social zones by gender (top part). The same data may be further explored by requesting more details on patients e.g. by revealing additional social zones by age group.
Further intuitive exploration of the data set may be facilitated by simple visual querying mechanism to reveal additional detail.
Extensions, relative benefits and value The above scenarios are only illustrative of what social networking can bring to eCancer practices such as research trials and open access publications. Moreover, the design concepts presented serve only the purpose of illustrating how tagging and other Web 2.0 technologies could be explored to this end. Obviously, as in other social networking applications the actual value is to be assessed only through time and by allowing users (of various types) to become engaged and provide contributions. Therefore, one could imagine various extensions of the scenarios sketched above so as to allow:
- Authors to revisit and change their data (without loosing earlier versions)
- Third parties to dispute basic conclusions in a study or
- Third parties doing similar things to inform the community of new sample data sets and invite commenting
- Patients to pose a relevant question
- Funding bodies to specify areas on research relevant to the drug
- etc
Living guidelines Reinterpreting the scenario presented in section 0 we could easily imagine a similar approach being applied and validated for guidelines. This time the documented material and its supporting evidence should be organized to reveal the content of a guideline and how it evolves in the course of time as new evidence (for and/or against) is accumulated.
Patient communities Patient communities can also benefit from a more informative exploitation of data as discussed in section 0. In what follows we present (hypothetical) examples of patients access to such repositories and the added value resulting from extracting information that reflects a personal profile or a point of view. It is important to note that our emphasis is not on typical community support functions (i.e. posting requests, receiving comments, sharing perspectives, etc), which are more or less available in social networking sites on health. Instead, we seek to understand how eCancer practices can be translated into meaningful and useful information in the users language. To this effect, advanced querying techniques can provide an important mechanism for disseminating research outcomes to the public as well as for classifying and presenting search results in a profile-relevant manner. Once gain our elaboration is purely exploratory and seeks to convey the concept rather than describe a particular solution.
Search = dynamic querying + exploration
Searching can take advantage from appropriate visual representations to present results in more intuitive manners. Let us imagine a patient interested in finding information broadly relevant to a medical profile. With current search engines, the way to approach the task is through keyword search. However, patients would appreciate presentation of search results in clustered collections or social zones of preference. Assuming some sort of mechanism / profiling technique whereby patients codify a social zone of preference, this presents the results of a Google search (the example depicts the search condition cancer research trials) in two different ways. The visualization presents the results by creating as many zones as the chosen classification scheme (i.e. number of different types of results such as web, video, images, books, etc.). Each result is presented as a node within a social zone and the size of the node is determined by the degree of interest, which in this case is the rank of the result within the data set.
Alternatively the same data are presented using a data mountain layout (i.e. nodes in the back of the graph get smaller indicating distance from the point of the users focus). This allows users to define the degree of interest by direct manipulation of the desire node. Furthermore, it would be possible to imagine scenarios of re-calculating the layout by changing / modifying the perspective of relevance.
Steps towards the solution
We have described preliminary results indicating how social media and Web2.0 technologies could be used in the context of EuroCancerComs to promote social networking within eCancer communities. The key idea behind the concepts is that users incrementally become content providers either by reflecting upon available knowledge or by establishing networking patterns. A critical technology is tagging which enhances the way of looking at a document, a web page, a speech, etc. Combining tagging with advanced and dynamic querying mechanisms is one way of engaging a variety of users in meaningful and recurrent exchanges in virtual space.
In light of the above, our vision of the ecancerHub social networking system can be summarized as follows:
Ultimate target: Design a virtual space to allow social networking on selected scenarios relevant to eCancer.
Context of operation: Social web site independent of existing portals (much like Google or Facebook) explicitly tasked to facilitate an emergent knowledge management process on eCancer digital content (various types and formats)
Properties of the social web: These may include:
Visualizing collective intelligence on eCancer
Replace currently prevailing form-based search with visual and exploration-based querying techniques
Interoperability with other systems on the grounds of widely accepted protocols and web services
To build such a system, we need to agree on the following:
Relevant information product families
What exactly is to be codified and how
Data types and formats to be supported
Indexing schemes, classifications, tagging
Flexible displays (plastic and role-based)
Orientation
Navigation
Exploration-based querying
Direct manipulation
Conjunctive versus disjunctive queries
Social aspects
Comments / annotations / voting / ranking
Negotiation / argumentation
Digital histories
The above could be tackled in a technical workshop aiming to address:
Community management
User roles, registration and access to information
Electronic neighborhoods (i.e. patients, professionals, mass media, funding bodies, etc)
Communication / notification services (threaded discussions moderation, asynchronous notification services)
Information content and interoperation with existing data bases and portals (e.g. dedicated portlets may be needed to assemble data from distributed sources and present them in a concise way; these portlets may be different if the material to be assembled is a paper and its resources from other types of medical data)
Examination of a suitable enterprise portal to serve as the community support system
eCancer practice
what practices are to be supported (i.e. cancer research trials, open access publication, etc)
for each practice we will need to define
- who is entitled to engage in a designated practice i.e. cancer research trials
- what processes are to be supported (i.e. registering the trial, reporting trials, conducting trials)
- indexing schemes and classifications (i.e. type of trial, drug, treatment / intervention, trial status / phase, sponsor, etc)
etc
Who does what, when and how (in a reasonable time schedule
TECHNOLOGY / ARCHITECURE
Text mining: developing and applying novel methodologies to cancer science information
The scienti?c literature provides descriptions of biological discoveries characterizing cancer at different levels of granularity. At the molecular level, articles contain detailed biochemical characterizations of genes and proteins in terms of sequence, structural and functional features. To detect associations between particular genes and diseases or phenotypic properties related to cancer, considerable amounts of population, epidemiologic, or even gene knockout studies using animal models have been published. The literature plays a central role in the current scienti?c discovery process, from the initial stage of experiment planning to the ?nal step of result interpretation and the subsequent generation of new hypotheses. Interpretation of genetic alterations observed in cancer patients using information hidden in the vast volume of scientific articles is increasingly challenging.
Eurocancercoms was able to map, describe and integrate the state of the art in the application of text-mining techniques to understand the cancer disease. It also described two novel text-mining methodologies developed and implemented by the consortium to discover genes that may directly implicated in malignant melanoma, and which are flexible enough to be adapted to other cancer types.
Based on the melanomas section of the National Cancer Institute (NCI) website a query to search PubMed was built. The query combined several MESH terms: melanoma[majr] AND human [mh] AND english[la]) OR (melanoma*[ti]). We restricted the search to articles published in the last ten years. As a result we obtained a list of 15,525 melanoma articles.
The second step filtered out those articles that do not contain molecular information in their abstracts. Many of the articles collected in the first step included topics like surgery, treatments guidelines, etc. that doesnt bring any light about the genes that are possible involved in this disease. The ABNER software was used to tag the bio-entities (protein, DNA and RNA mentions) in the abstracts automatically, retaining only those that contained mentions of these bioentities. After the filtering process, 4,580 articles tagged with their corresponding genes and proteins were recovered. The resulting set constituted the positive training data for the subsequent semi-supervised machine learning based document classifier. In order to define a negative (non-relevant) training set, a random list of 4580 papers (balanced data set) containing genes or proteins in their abstracts but not directly associated to melanoma was created.
Pipeline that summarizes the novel method developed by eurocancercoms consortium to rank the genes according to their association with melanoma. To classify the documents we used a Support Vector Machine (SVM) with a linear kernel function using the SVMLight and LibSVM packages. The classifier was trained to generated a predictive model and validated through five cross-validation.
In the second approach a list of melanoma genes extracted from the Biocards section of the MMMP database was extracted. The goal in this case was to find a possible functional interpretation of the genes based on literature analysis. SENT was used which is a functional interpretation tool based on literature analysis. SENT uses Non-negative Matrix Factorization (NMF) to identify topics in the scientific articles related to a collection of genes or their products, and uses them to group and summarize these genes.
SENT clustered the genes in 8 groups, for each group the literature terms, the genes names and the number of articles supporting the decision where displayed. These results provide a general idea of the topics latent in the literature for the MMMP genes.
The results of the two strategies followed by the eurocancercoms consortium to utilise text-mining methods to understand the relation between genes and the melanoma disease provided the first novel scientific approach to integrating science communication to enhance dissemination and utility of the published literature thus providing the first proof of principle aimed at addressing info-overload.
Architectural approaches to cancer science information: analysis & integration
Eurocancercoms has been continuously working on an application that allows the research community to explore the cancer disease from an integrated point of view. Our aim was to take advance of the revolution in the analysis of genetic diversity in humans and its role in disease that has come with the advent of massive parallel sequencing (MPS) strategies. These methods allow the acquisition of genetic information on a massive scale (at a cost of somewhat reduced accuracy) and low price.
New large-scale genomics efforts are generating data at unprecedented scale. The 1000 Genomes Project (www.1000genomes.org) alone is expected to produce 60 times more data than has been deposited in the public repositories over the last 25 years, and has already surpassed the amount of sequence data in Genbank. The information generated by projects such as The 1000 Genomes Project will allow us to find relevant variants for different cancer types.
In a first iteration somatic mutations were extracted from the COSMIC database and germ-line mutations from dbSNP. For each gene the application displays the number of germline and somatic mutations found, and details about each mutation.
In this new version we added information about pathways extracted from the KEGG database (http://www.genome.jp/kegg/). It has been proposed that the genes mutated in cancer tend to encode proteins that cluster in certain signalling pathways, and different sets of genes related to specific tumours might, in fact, correspond to a unified set of affected pathways or functional processes. This pathway-mapping approach has been applied in order to discover pathways that may be particularly relevant in melanoma.
Currently the application provides also Information about drugs that may be used to treat the disease. We collected the information from NCI (http://www.cancer.gov/) PharmaGKB (http://www.pharmgkb.org/) and Matador (http://matador.embl.de/). The relation between a drug and a gene may be a direct or indirect. In the first case the gene is a direct target for the drug. Indirect relations are established based on information derived from literature or from pathways analysis. We are currently working on categorizing the information about drugs to offer a more complete scenario to understand the drug activity.
Finally we also introduced the results generated by the two text-mining approaches described before, as a way to understand the possible implications of the genes in melanoma. For each gene we provide the rank obtained from the SVM classifier and also the terms obtained in the clustering analysis.
The application is available at: http://eurocancercoms.bioinfo.cnio.es. The main window contains a table with the MMMP gene list and the information about mutations, pathways, drugs, cancer types where the gene plays a relevant role and the literature score provided by the SVM classifier.
Proto-type for e-engagement for prevention (nutrition) and R&D
Nutritional status is an important factor in prediction of the risks associated with cancer disease and treatment affecting clinical outcomes, including survival and the ability to tolerate treatments. Nutrition management has been proposed as auspicious to cancer patients because the nutritional status and the nutritional intake are relevant factors affecting also the quality of life and the psychological status of cancer patients. Nutritional screening has been identified for early intervention in clinical practices, however there is not always proactive strategy that can be used when the patient is discharged to home. The number of cancer survivors continues to increase due to many factors, including a growing aging population, early detection, improved diagnostic methods, more effective treatment, and improved clinical follow-up after treatment. It is important for physicians, patients and public health professionals to be knowledgeable of issues survivors may face, especially effects of treatment on their physical and psychosocial well-being. This understanding is critical in promoting good health and coordinating comprehensive care for cancer survivors. Furthermore, a significant proportion of the worldwide burden of cancer could be prevented through the application of existing cancer control knowledge, implementing programs and public health campaigns promoting, for instance, physical activity and healthier dietary patterns.
The world wide web has changes the mass and health communication in the last decades. The Internet and the associated technologies can contribute to disease prevention and health promotion in a number of contexts, enabling a swift, inexpensive and easy distribution of content information; this may turn experience into learning. The Internet offers potential for interaction and is a medium for influencing learning and behaviour change. Evidences from randomized studies suggest support through Internet to cancer survivors in order to reach a positive outcome regarding social functioning, ability in searching for information, depression and state of health perceived though cancer patients quality of life seems not to have improved.
This S&T result was derived from a collaboration between eurocancercoms and non-eurcancerom partners: the Organisation of European Cancer Institutes (OECI), the Italian Association for cancer patients, their families and friends (AIMaC - Associazione Italiana dei Malati di Cancro) and the Italian Federation of Volunteer-Based Cancer Organizations (FAVO - Federazione italiana delle Associazioni di Volontariato di Oncologia).
The change in quality of life, psychological status and nutritional knowledge, measured by study questionnaires was used to evaluate the effectiveness of the prevention approach based on the 6-month visits of a website dedicated to study. This novel architecture - Internet space where tItalian patients could find information about nutritional care and where to interact (forum and blog), was shown to positively improve knowledge, quality of life and psychological status of the oncological subjects.
Integrating cancer communication & information: a demonstrator social media platform ecancerHub
Opus: Core platform
Opus is Reality Digitals tried and tested video and community and social media platform and comes as standard pre-configured with the following functionality:
Secure registration
Secure login
Single sign on
Profile creation
Blogs
Forums
Friends
Groups
Google Map integration
Video ,Audio, Graphic and Text upload via:
- Web
- Webcam
- Mobile
- Camera
- Bulk upload
Content syndication too: Youtube/Facebook/Micro-site (if required)
Comments
Ratings
Industry leading moderation dashboard
Extensive administration and auditing dashboard
eCancer Community functionality
Homepage
Header Displayed on all pages
Logo
Lead site sponsor logo
Leader board advertising banner
Navigation bar (Standard Opus functionality) displayed on all pages
Home
Articles
People (members)
Groups
Videos
Blogs
Chat and Answers
My eCancer Hub (only viewable after member login) Prompt to login
Site search
Homepage content sections: Editorial (left hand iFrame)
Content box Welcome copy
Content box Latest on eCancer Hub (list of editorial topics from publishing partner)
Content updated automatically via RSS or XML
Content box Latest Chat topics
Automatic Pull from chat section - Opus
Content box Video wall (Member and Professional content)
Automatic Pull from video section
Content box - Featured members
Automatic pull from member profile section
Permanent commercial content displayed on all pages (right hand iFrame)
Content box Login or register section
MPU and Sky Scrapper ad spaces
Market research polling functionality
Content box Useful links section
Content box Sponsored section
Content box Latest News
Content box Site sponsors
Registration Page Opening in left hand iFrame
Blank box User name
Blank box Name
Blank box Surname
Blank box Password
Blank box - Email address
Drop down Country
Blank box Postcode
Drop down User group
Large blank text box Tell us about yourself
Check box Send weekly news letter
Button Enter
Email notification of accepted membership, auto generated and sent to new member
As a member youll also be able to:
Write or read blogs
Ask or answer queries in our chat forum or have a good old moan!
Comment on any articles you read
Post your videos into our galleries
Meet a huge network of cancer professionals
Get advice
To bookmark ECancer Hub (so you never have to search for us) click here.
Profile page Accessed though My ECancer Hub button on the Nav bar.
Profile tab
As registration page functionality +
Upload picture
Save and edit button, so you can edit Tell us about yourself text box
Navigation tabs on top of Profile page (next too Profile tab)
My Friends
Displays current friends and pending approval friends
Displays total number of friends
Button Find new friends
Find new friends opens the People/member page
Click on a member and view their profile
Button Invite as friend
Opens text box write a message to person inviting them to be a friend
Button Send invite
My Groups
Displays current subscribed groups and group invitations
Button Create a group
Opens Create a Group page
Text box - Group name
Text box Describe your group
Drop down Select your subject
Check box Private
Check box Public
Button - Create
My photos and videos
Displays uploaded photos and videos
Button Add new photos or videos
Opens upload media page
Text box Browse for media (on your PC)
Text box Name your photo or video
Text box Date taken
Check boxes Who should be able to view this
Public, Private, Friends
Drop down Publish to gallery
Text box Tell us all about it
Button Upload
Button Cancel
Button Add another
Articles Page- Accessed through either the Top navigation bar or by clicking on an article in the Latest from ECancer Hub section on the homepage
Opens into the left hand iFrame on the page.
Lists out the latest stories/articles on eCancer Hub
Picture graphic and brief copy displayed for each story or article
Content is auto feed from publisher via RSS or XML
Click on story/article
Full story or article opens into the left hand iFrame
Potential hyper links within the story highlighted in blue
Hyper links open into left hand iFrame
Articles can be sorted by: Most recent or most read
Button Back (cycles you back to main articles page)
Reporting
ECancer Hub will be provided with 24x7 access into all of Reality Digitals Opus reporting tools.
ECancer Hub will be able to monitor all site activity and capture critical metrics in real time.
ECancer Hub will be provided administration tools to QA clips/pictures, comments and ratings before publishing: Ensures quality and content standards are met before publishing
Omniture or Google tags are easily added for additional reporting depth (upon request)
Reporting Metrics Include:
File List - Detailed list of all media files on the site
User List - Detailed list of all users and registration information
Group List - Detailed list of all groups and activity
Top List - List of the top 100 media files by date, number of views, ratings or votes
Daily Report- Daily activity report of number of users, registrations, uploads, media views, comments, ratings, etc.
Encoder Status - Real-time status of all clips being encoded
Web Logs - Reports on server activity/traffic
Bandwidth Report - Daily and monthly reports on bandwidth usage
Moderation Tools:
The site administrators have access to a robust set of moderation tools via the Opus Dash¬board, which allows for monitoring and reviewing of User submissions before pub¬lishing to the site and appropriate actions. Moderation tools include:
Pending List - Detailed list of all media awaiting approval with approve, re¬ject, and delete functions
Blog Admin - Administer all blogs on the site
Comment Report - Detailed list of all comments and approval functions
Blocked IPs - Configuration of IP addresses banned from the site
At the user-level, users have the ability to flag content as inappropriate.
Filtering
The administrator can identify a list of offensive words or phrases that prompt fur¬ther action or automatic deletion. (English only unless customer provides international word list)
Video Branding
ECancer Hub videos can be branded with pre-roll and post-roll video which stays with the video wherever it is distributed.
SEO
Search engine optimization (SEO) of content will be achieved through HTML meta-tags and alternate content
Hosting Support and CDN
The entire ECancer Hub program will be hosted and supported by Reality Digital. SAN storage systems are monitored 24x7 and are fully redundant and backed up hourly to NAS. Daily tapes and backed up and removed from site.
Reality Digital provides system level support across the entire solution including Apps, backend and CDN. Customers are provided with all relevant contact information.
For DHP we will leverage the Akamai CDN to dramatically improve User experience from most anywhere in the UK and for than matter on the planet.
Storage & Search
Reality Digital Opus is based on industry standards such as Microsoft COM and ASP, SOAP, XML, HTTP and SMTP. The content in the database can be easily exported into an XML without impacting any existing systems.
Storage is architected on a redundant fiber connected SANs replicated to a NAS for additional redundancy. Additionally, Flash media is populated on the CDN at the time of ingest and encode.
Reality Digital offers SQL based search. The users can search the content using keywords and metadata and a user by the username, name or email address. Optionally, we can integrate with any 3rd party Search Engine to support more advanced search features. We have experience with most major search applications. The addition of one of these search engines adds significant expense, but all rich capabilities for search, recommendation, and highly customized user experiences.
Data Protection
1. Reality Digital powers all sites on redundant database servers to ensure availability
2. RAID 5 SANs continually replicated to NAS devices for back-up of all data and media
3. Nightly backup of servers
4. Nightly backup of data to tape Weekly rotation of tape is stored offsite
5. Virtual clones of web servers allow us to bring up a replacement within minutes
6. In some cases, where load balancing is part of the configuration, we have several identically configured servers standing by to prevent down time
7. Redundant Switches with redundant uplinks ensure constant connectivity to the internet
8. Redundant Encoding and Ingest services provide fast availability of incoming media
9. Limelight and Akamai CDN integration provides additional data redundancy of content
Third Party Integration: integration with many third party technologies including: Adobe, Akamai, Doubleclick, Google Maps, Limelight, Microsoft, Omniture, Paypal, Word¬press, and many more, including: eCommerce and retail customers.
Additionally through the template development work with TME, integration of the event registration and booking systems should be straight forward.
Professional Services and Program Delivery
Because the Reality Digital platform has already been built out, the actual development team required is very small. A typical project requires a project manager, a project architect, a designer (as required) and a QA team. Some projects demand a second engineer and a Flash developer as well. Because we build social video sites day in and day out, a typical project will take from 6 8 weeks all in to deliver.
Potential Impact:
The project has been distributed into the public domain through two urls:
1. www.eurocancercoms.eu/
2. http://www.ecancerhub.eu/
In addition there have been numerous non-publication e-distribution resources. For example,
Engaging the patient community: round-table and media distribution
Engagement using You-Tube
Extensive e-banner and distributed web-site advertising and PR
ecancerHub: a platform for developing novel cancer science information tools
Impact and dissemination are listed in following pages.
List of Websites:
1. www.eurocancercoms.eu/
2. http://www.ecancerhub.eu/
The lack of efficient communication among cancer health professionals, patients and policy makers remains a significant barrier to collaboration in the EU (EUROCAN PLUS Report to the European Parliament, 2008). Information overload and a very fractionated, exhaustive array of resources, networks and knowledge providers are seriously hindering the translation and application of research findings in Europe. With the constant explosion of data the cancer community can expect to face increasingly challenging times for reliable and effective scientific communication.
The aim of eurocancercoms was to establish an integrated model for a Europe-wide cancer information and policy exchange portal that would provide a functional exchange system for accurate information and intelligence, catering to the needs of health professionals, patients and policy makers. Such a model could then be subsequently applied to other areas of healthcare.
A consortium of the major European cancer community organisations in collaboration with e-communication technical experts carried out major and systematic studies of all aspects of e-cancer information space, including analysis of novel architecture, applications and tools. Surveys of the various cancer communities patients, healthcare professionals with particular foci, and novel content including specific projects on cancer clinical guidelines. The project also produced an array of major policy studies and reports into all facets of cancer communication, from open access to the state of R&D communication within academic cancer surgery (a major but orphan area).
These studies, surveys and analysis were brought together to develop and launch a novel social media platform social to EU cancer communications eCancerHub (http://www.ecancerhub.eu/) which has served as space for bringing together 4 different communities patients, healthcare professionals, scientists and policy-makers.
Project Context and Objectives:
The lack of efficient communication among cancer scientists, health professionals, patients and policy makers remains a significant barrier to collaboration in the European Union (EU). The recently completed EUROCAN PLUS project surveyed all constituents in Europe involved in some way or another with the cancer research business and concluded unanimously that information overload and an overwhelmingly fractionated, exhaustive array of resources, networks and knowledge providers are seriously hindering the translation, application and implementation of research across Europe. With the constant explosion of data we can expect to face increasingly challenging times for co-ordinated, reliable and effective scientific dissemination and communication.
The EU needs to establish an integrated model for a Europe-wide comprehensive cancer information and policy exchange portal. Such a model could subsequently be applied to other areas of healthcare, such as cardiology, neurobiology etc.
Cancer is a good model as its science is far advanced, subspecialisation within basic and clinical research has splintered the area into hundreds of boxes, and its impact on the health of Europeans is a significant issue. As the population ages, deaths due to cancer are expected to rise from around one million to almost double within a quarter of a century. Worldwide, the WHO has now declared that cancer causes more deaths globally than malaria, AIDs and tuberculosis added together. There is at present no workable model centralised or networked which even begins to address the communication challenges within this area of health science.
To address this void, the consortium will conduct an inventory of all existing information tools currently in use, including gene databases and the NCI resources in caBIG, faults and flaws and requirements for the future. E-surveys and other online research tools will be used to provide rapid and easily collated information. This will include collection of current national and international legal and regulatory requirements, for instance on Data Protection, Privacy and Ethical Issues and the provision of strategic intelligence on cancer research activities for policy makers, including those in Pharma and biotech companies. A review of new technologies available to assist in the dissemination of the information will be carried out. The consortium will seek to define and establish a state of the art communications system to connect and involve all those implicated in translation of basic cancer research into clinically testable hypotheses, public health prevention and management strategies, patient information, participation and support activities
The European Cancer Research Managers (ECRM) have been set up by the EC to conduct research of different populations in the cancer arena within Europe. They will be responsible for all e-questionnaire design and analysis, exploiting their partnership with the University College, London School of Library Archive and Information Studies SLAIS.
One of the coordinating partners, ecancer.eu will act as the medium for testing and feedback, and then will be used as the testbed for technological solutions to the problems unveiled by the consortium. Although ecancer.eu has state of the art web and TV capacity now, by the end of the project it will have to have adopted the latest technical solutions emerging from academic and industry sources. The proposed project will involve representatives from a wide range of scientific and clinical societies and disciplines (ECCO, EORTC, EACR, EUSOMA, SIOPE, OECI and ESMO). These organisations represent over 50,000 cancer scientists and clinicians from across Europe spanning molecular analysis of cancer cells to palliative care in adults and children with cancer. They will work to define the barriers, problems and requirements in publishing in a field of science which has exploded in its information content and multiplied its dissemination challenge. But the study will go further, to examine post publication information dissemination, understanding by patients. Tenovus will lead the patient (consumer) research in areas such as requirements for believable information, understandable language, and communal use and mutual sharing of support mechanisms, generated by the cascaded information in the project. Also, in the consortium are highly experienced academics and practitioners of the science and technology of communication particularly e-health, e-genetics, human computer interaction design, virtual communities, electronic villages, digital active television (Brandcast, FORTH, SIVECO and CTRC). Two global technology companies (GE Healthcare and Philips) are partners and they have immense depth in communication solutions, particularly in imaging, and each specialises in oncology.
The European Society of Medical Oncology (ESMO) will facilitate debate and provide expertise relating to communications between health professionals in oncology, consumers (cancer patients) and the pharmaceutical industry.
There are two proposed supporting co-ordinators. ECCO, the European CanCer Organisation, consists of all the principle cancer societies, from basic research to clinical care delivery, and its President Professor Lex Eggermont will coordinate the Working Parties, charged with specific Work Packages designed to inventarise scientist, clinician, pharma and consumer usage, and bottlenecks to more efficient communication. The second supporting coordinator is ecancer.eu chaired by Professor Gordon McVie. He will organise the Working Parties consisting of technical experts in web use, digital television, grid techniques, library functions, Open Access, conventional publishers and European Policy experts. Ecancer.eu is a new online, free, open access cancer journal from the European Institute of Oncology (IEO), which already receives over 12000 visitors per month. All documents arising from the consortium will be published immediately on a dedicated part of the journal, and e-questionnaires will be organised through its website.
A Scientific Committee (SC), chaired by Professor Veronesi from IEO will coordinate the whole consortium, ensuring that bridging between demand and solution is efficient and speedy. The SC will analyse the reports generated to identify common resources to optimise the best way forward, to determine priorities for fast and efficient communication within cancer research in Europe, and to integrate the ideas into a common strategy and to supervise feasibility trials on ecancer.eu.
The Coordinating Action (CA) will make recommendations for the publishing and dissemination of cancer research in the EU including identifying further priorities for the 7th and even 8th Framework Programmes. These will then be widely circulated throughout Europe prior to finalisation by the SC and submission to the European Commission. Final reports will be published in ecancer.eu and conventional journals, such as Lancet, European Journal of Oncology and Annals of Oncology. Their publishers have declined to be partners, but remain willing to participate in workgroups and dissemination activities. The CA will, among others, address the following barriers to efficient information flow:
Basic research
The need for a scientist to be original and to publish their research first. A scientists career progresses according to the number and originality of papers they publish, they are therefore unwilling to share initial findings and remain secretive about ongoing data. Negative results are infrequently published in scientific journals [1].
Cancer networks will share information, though not often raw data, within their own network; however they rarely link with other cancer networks on a functional basis. New initiatives to publish raw data directly from scientists e-lab books will be studied
3. All new discoveries require a patent to secure the intellectual property. Usually a bio-tech company is founded to prove and exploit the patent. Once commercial interests are involved the release of information becomes timed for commercial advantage. Is this a significant barrier?
4. Pharmaceutical companies carry out a large proportion of research in Europe (around 50% in the cancer area), but 80% of their science findings are never published. Commercial interests, including stock market listings influence the publishing, and the timing of publication, of scientific research. Several major Pharma companies have indicated a strong desire to be involved in developing solutions to this issue, though not as full partners.
5. Language barriers among scientists severely hinder rapid science dissemination, as the vast majority of science publications are in English and the majority of European scientists do not have English as a first language.
Technical language barriers within subsets of the oncology community create communication barriers. Patients are also excluded from understanding new research by the use of unfamiliar technical terminology.
Differing regulations and procedures in EU member states prevent cancer research being disseminated across the EU. Procedures need to be standardised to allow for the free flow of research and this will require European coordination.
Peer Review is fast approaching its sell-by-date. What is to replace it?
Clinical Research
Clinical research has the added restriction of maintaining patient confidentiality. The issues of data protection and privacy are of prime importance to the consortium. Fortunately we have two partners from another EC programme ACGT, who have already far-advanced proposals to solve these issues. The Independent Ethical Committee will take a pro-active role in policing this topic.
Conflicts of interest are of potential importance when clinical trials report results on new molecules and devices from the pharmaceutical sector.
There is ample evidence which highlights the heterogeneity of outcomes of cancer care throughout Europe. What is clear from all of them is the detrimental impact of lower social class, deprivation, ethnicity, low educational achievement, and access to relevant information amongst cancer patients. These issues are seminal to this project and the communication needs arising from them will be investigated and analysed. A variety of appropriate solutions will be developed to meet the identified needs.
Patients are increasingly using the Internet in a hunt for reliable cancer information. One website on cancer opens each day of the year, and the word cancer is searched on Google more than any other except pornography! Confusion is rife, total failure to translate published results of research is all too common, and public understanding of the reality of cancer prevention messages and access to best care (including access to clinical trials) is abysmal. Needs of patients for particularly relevant information, its optimal timing and mode of transmission will be determined.
A particular area of missed opportunity is the communication of information to patients and the general public on risk. Cancer is a genetic disease, and lack of understanding of even the most important familial genes in breast, colon and ovarian cancer has led to failure of application of important scientific discovery. Two partners have particular expertise in this difficult territory (EACR and Tenovus), and again the ethical issues are numerous and must be solved.
An important aim for the Commission is to stimulate analysis and debate to improve the standards of the European science system and its ability to produce robust and reliable knowledge to support policy-making and respond to social needs.
Project Results:
COMMUNITY ANALYSIS
Paediatric oncology
The State of Research into Children with Cancer across Europe New Policies for a New Decade was published by the European Society for Paediatric Oncology (SIOPE) and ECRM Foundation. The report into the e-communication needs of the paediatric oncology community was presented at a special event in the European Parliament (wednesday 9 February) to mark International Childhood Cancer Awareness Day to a wide range of EU policymakers as well as young cancer patients and parents.
They found that countries with a larger oncology burden, such as those in Eastern Europe, tended not to collaborate in research with those with a better-developed research structure, and this in turn affected the care they were able to give young patients. Paediatric oncologists from ten countries were asked about their national situations and the differences were striking, said Professor Kathy Pritchard-Jones, Cancer Programme Director at the UCL Institute of Child Health, London, UK and Past President of SIOPE. For example, a respondent from Bosnia-Herzegovina cited lack of hospital space, no specialist paediatric oncology radiologists, and inadequate diagnostic facilities, with everything adapted to adult patients. The Czech Republic said that research grants were usually for only three years and there was little support for young scientists to attend international courses and congresses which could help harmonise the level of knowledge across European states. In Italy, it was the fragmentation of childhood cancer care which concerned doctors, with nearly 50 centres specialising in paediatric haematology and oncology, and a lack of co-ordination between research laboratories and clinics. In comparison, Sweden has only six paediatric oncology units, which work closely together but, even there, there are problems of funding and a lack of experienced staff. The UK expressed concern about the effects of health service cuts on childrens cancer care, and the fact that infrastructure changes had moved paediatrics alongside the adult cancer model, meaning that research for children now has to compete for funds at a much more visible level with research for adults.The report also found large differences in the provision of information on childhood cancer, with variations in the involvement of parental organisations, the use of digital media, and the adoption of a common national standard for information provision. When a child is sick, the provision of accurate information to the patient and family is essential said Professor Sullivan. We believe that the establishment of a European Common Information Portal could do much to tackle major deficiencies in information in countries with few or no patient organisations, or where the existence of a large number of languages make access to such information difficult. The report called for adequate long-term EU funding to support a Europe-wide clinical trials network for paediatric oncology. This is an essential prerequisite for the effective testing and dissemination of new therapies and techniques, said Professor Pritchard-Jones. We also need to study treatment outcomes, and for this the creation of a European Childhood Cancer Epidemiological Registry is essential. With these tools we can maintain the enormous progress that has been made in the past. Without them, we run the risk of jeopardising that progress and failing children who are dependent on us for their survival.
[*Sullivan R, Pritchard-Jones K, Camporesi S, Lewison G, Vassal G, Ladenstein R, et al. (2011) The state of research into children with cancer across Europe: new policies for a new decade ecancer Vol 5 doi: 10.3332/ecancer.2011.210]
Cancer Surgery
Eurocancercoms made extensive use of bibliometrics to map out the various cancer communities operating in R&D. Here it analysed global cancer surgery R&D trends over the last ten years across 21 countries.Surgery is the main modality for cancer cure and control globally. Yet, in comparison to other areas such as cancer drugs, we know little about ongoing research activities to inform policymakers. Two sub-field filters, surgery research and oncology, were developed and applied to Web of Science. The intersection of these two filters identified papers in surgical oncology, and their bibliographic details were downloaded for analysis. This included matching of five-year citation counts to the papers, impact factor, geographical analysis by country, translational collaboration, involvement in clinical trials, citation on clinical guidelines, and percentage of reviews. Surgical oncology represents about 9% of all cancer research low in comparison with surgerys contribution to cancer treatment. The USA published the most, followed by Japan which had a high relative commitment to surgery within cancer research, followed by the large West European countries. Although Swedens papers were relatively basic it participated the most in clinical trials. Its papers were also the most cited on clinical guidelines, but contained relatively few reviews, where the UK, Greece and Belgium scored best. Surgical oncology papers are generally not well cited compared with cancer research overall, but on this measure the Netherlands, the USA and Sweden scored best. International collaboration was measured relative to what might have been expected, on this indicator Canada, Switzerland and the USA were the best performers. The results generated enabled a) new policy intelligence to be created and b) provided the spine for e-contact and networking between the active academic cancer surgery community through ecancerHub
Patients & Risk
Risk in cancercommunication: The main findings are presented as the key themes and issues arising from the RTD. There was consistency in the group around the need for improvements to the risk communication process, particularly around information and its onward diffusion i.e. Telling the family and differences regarding wanting to know their genetic status. This was evident from the responses and the more extensive discussion on these topics. The women shared their perspectives and experiences of breast cancer or risk communication and those of others in similar situations from their personal or advocacy group connections. The additional Health Care Professional perspective contributed examples of research and their knowledge of services relevant to the discussion. The richness of the data emerges from the group dynamic and their diversity as not everyone had the same views and experiences due possibly to differences in, for example age, education, and in particular in this situation, across countries.
The following themes were identified from the RTD. Key quotations are included in the text to illustrate the theme and to show that within the topic there exists a range of comments, sometimes opposing.
The participants are identified as HCP1, HCP2 (health care professionals), P3-6 (patient representatives) where P4 added comments pre or post interview = 4a and 4b, R7, R8 (Macmillan Cancer Voice reviewers)
The Round Table Discussion was introduced with a brief overview of cancer genetic services so this not only set the scene but provoked some discussion about current service provision and recommendations for potential improvements. A person entering the cancer genetic service was described as there for life due to the changing information and support needs at different stages through the journey. HCP2
There were several positive comments about the service, being non directive and allowing sufficient time to discuss things, but disappointment with lack of support when passing on the news to family members:
Twelve years ago you just got a letter with your diagnosis and to tell your family about it and Ive heard from other people its still the same P3
Involving patients in service redesign was advocated including involvement in guidelines.
How is the whole process of communication experienced by people who have a predisposition, what do they need, what have they missed and what solutions are proposed P4a
persuading, ...professionals not just to listen to women but actually to convene groups of people and to try and be informed.... looking at how populations get information and how they interpret general information...if that were pursued with more vigor certainly in Ireland I think wed have better outcomes. P5
There was recognition that a key beneficial difference with specific cancer genetic services compared to consultations with general practitioners (GP) was the planned time for discussion.
not like your six minutes with your GP in the UK HCP2
Media
From the 240 media contacts contacted to complete the survey, there were 63 responses (26.25% response rate). The respondents were first asked to give their occupation:
The majority of respondents were from the United Kingdom (42.37%), followed by the United States (18.64%). The remaining respondents came from Australia, Belgium, Finland, France, Germany, India, Italy, The Netherlands, Spain, Canada and Asia/Pacific.
The professional experience of respondents weighted heavily towards those with more rather than less experience: 53% had 20+ years of news/media experience, 12% 15-20 years, 18% 10-15 years and 17% had less than 10 years experience.
Respondents were asked in what type of news organisation they worked, with multi answers being allowed. There was representation across the all categories given:
Respondents were then asked to select their areas of focus. The large majority (71%) selected healthcare, followed by science stories related to topical news and events (47%), 31% said specific scientific subjects, 19% general science stories and 10% science events. The remaining areas of economy, general news, industry and finance accounted for less than 11%.
Respondents were asked to rank what influenced their selection of science topics. 50% ranked public interest/opinion as the highest influencing factor, followed by topicality. On the other end of the scale, industry involvement (32%) was ranked as least influential, followed by economy involved topics and organisational press releases (both 21%).
The survey also asked respondents to indicate how interested their readers/viewers are in a selection of topics. Topics around patient involvement and stories specific to cancer type are of most interest (70% and 74% respectively). Readers are also more interested in topics such as drug development, genetics, prevention and new technology than not. Local stories, however, were said to be of very little interest (49%). Respondents were then asked to state how interested in their editors/managers are in the same topics. These responses almost evenly matched those from the previous question, with patient involvement, stories specific to cancer type, and drug development being just as interesting to the Editors.
Respondents were given a list of potential problems that they may come across when reporting science news, and were asked to state how often they come across each. The following five problems were the most often/regularly encountered (in rank order):
1. Research findings overstated by media
2. Research findings overstated by PR people
3. Uncertain or ambiguous results
4. Research findings overstated by researchers
5. Technical jargon makes the news hard to find.
They were then asked how much they agree with a set of statements relating to experiences they face as reporters. The following three statements gave the strongest responses:
The fact that 65% of respondents agree/strongly agree to the idea You would appreciate a forum where you could access a pool of experts willing to comment/give quotes for your story validates the 60% of strongly agree/agree responses to the statement You long for one central source of trustworthy information. Respondents want a pool of sources rather than just one.
This is also supported by the responses when respondents were asked how they would prefer to receive potential news stories. Email was by far the most preferred means with 95% of respondents stating it as their preference. However, other sources of news delivery received a good spread of preference:
These results would suggest that if research organisations and scientists want their news to receive the highest impact in the media, then the stories have to be distributed to the whole spectrum of news outlets.
From a list, respondents were asked to tick all statements that they identified as being frustrations. In rank order, they were:
1. Press releases/sources are inconsistent in quality and reliability
2. Cited sources/PR people respond to slowly to your queries and you miss the opportunity to publish
3. Hard to distinguish really promising news among all the dross you receive
4. Scientists/researchers/PR people dont understand what interests the public
5. Scientists/researchers/PR people dont understand your story interests
6. Press releases or other stories are too complicated or technical
CONTENT ANALYSIS
Integrating cancer clinical guidelines: understanding current information provision
In total 30 European organisations in oncology were included in the survey. Of these, 21*/30 responded to the questionnaire, the majority of which represent specific medical specialties in oncology (38.1%) or disease specific organisations in oncology (23.8%). Other participating entities included an oncology pharmacy organisation, a general cancer society, an educational organisation, a nursing and a patient society. The majority of survey participants were involved in guideline development (62%), the percentage of which rose to 75% when considering the medical oncology specialty and disease specific societies alone.
The results of the survey revealed a number of serious pitfalls in the ways in which European societies are developing guidelines:
Notably, 31 % of organisations surveyed stated that they do not have guidelines or a structured process for guideline development. This also shows a lack of homogeneity in the methodologies to develop clinical guidelines. In addition, for the majority of organisations (69%), there is no methodological training for members of guideline development groups meaning that guideline developers are not specifically trained to do so.
Searches via electronic databases are the preferred method for the collection of evidence. For almost half of the guidelines developed original articles are confirmed manually. For the majority of the guidelines (85%) systematic reviews are the preferred method of analysing evidence on which a guideline is based. Experienced based evidence accounts for 69 %.
Several methods are employed in formulating guidelines. Both informal and formal expert consensus are equally adopted in approximately half of the guidelines development. An evidence-linked system weighing the evidence according to rating criteria is used in 38% of cases and a subjective review in almost a quarter of the guidelines.
The majority of guidelines (70%) are reviewed both internally and externally, compared with guidelines developed by other groups (61.5%), and endorsed by formal authorisation by a professional organisation of the target users (46.2%) the remaining guidelines are approved by the respective Societys Board. Also noteworthy is the fact that more than half of the guidelines are not piloted among end users (54%).
Importantly, the survey reveals that 62% of the guidelines do not have an expiry date. Although most of the organisations are regularly updating their guidelines, there is significant heterogeneity in the frequency of these updates which can vary from every 1, 2 or 3 years or when important scientific information becomes available. This shows that there is no systematic approach towards updating guidelines.
Most European Societies surveyed (77%), are systematically monitoring changes in the literature related to their guidelines. The regular quality systems used for the guideline program are developing and publishing criteria for good guideline development ('guidelines for guidelines') (38.6%); revision of the guidelines based on comments from the professional community (30.8%); appraisal of existing guidelines (30.8%); submission of the guidelines to a guideline clearinghouse; publication of the guidelines in a European journal and review and updating of the guidelines by an expert panel.
Another important outcome of the survey is that patients are often, in more than one third of the cases (38%), not involved in the development of guidelines. When they are involved, it is almost equally either through surveys of patients views and preferences; or reviews by representatives of patient organisations; or by participation in the guideline development group. The patients that are involved in guideline development groups are mostly selected by the relevant patient organisations (46.2%); to a lesser degree in existing patient panels (15.4%) and in a few cases through clinical centres (5,4%). None of the European societies surveyed use open advertised calls.
The majority of the guidelines (85%) developed are treatment and disease-management related. Half of them are related to diagnosis while only a minority (15%) are related to prevention or screening. In addition to ensuring appropriate clinical care which is the purpose of three quarters of the guidelines, for almost a quarter of them, the aim is also to seek cost containment. More than half of the guidelines are targeting policy makers and not solely health professionals who belong to the institutions that are developing the guidelines.
Recommendations/conclusions ..
Guidelines are essential tools for oncology professionals and patients alike. By following up to date guidelines, prevention (primary, secondary and tertiary), diagnostic procedures in cancer treatment and care can be carried out in the best possible way. As the survey demonstrates, guidelines are developed by many European cancer organisations. Since such development is both costly and time consuming for societies and their members, a more coordinated approach should be encouraged to provide up-to-date European cancer guidelines in the most cost effective and timely means possible.
Having provided an inventory of European groups developing guidelines and assessed the corresponding methodologies and approaches to guideline development based on literature and the outcomes of the survey, the following observations and recommendations can be made:
From the very outset the entity/entities developing guidelines should ensure an optimal environment for guideline development, systematically describe the various steps involved and provide a structured budget for all aspects including: methodological training of personnel involved in guideline development and subsequent evaluation; open access to specific data relating to the guidelines (full text publications); a detailed implementation plan; rigorous evaluation system and tool/s; quality assurance plan by which the impact of guidelines in clinical practice can be evaluated, as well as an easily accessible and interactive communications tool connecting all members involved in the process.
Importantly, both the focus and the stakeholders for which the guidelines are being developed must be clearly defined with greater emphasis on patient access and involvement in particular, and upon development/ revision(s) guidelines should be easily accessible for end users through a range of electronic tools and channels.
A workshop through which to formulate and establish best practice in guideline development would be favourable in order to improve guideline development across Europe. Further, since cancer related guidelines are becoming increasingly inter-and multidisciplinary, ECCO could play a key role in coordinating these efforts as the only multidisciplinary and multiprofessional cancer organisation that connects and responds to all stakeholders in oncology in Europe.
What information should ecancerHub integrate and how should quality be maintained?
An online survey of the cancer science community in Europe was conducted to discover their experiences of and attitudes to the issues surrounding academic publishing and Open Access. The responses from the two major surveys showed very positive attitudes to the Open Access publishing route; perhaps the most challenging statistic from the EACR survey is that 88% of respondents believe that publicly funded research should be made available to be read and used without access barriers.
Cancer practice guidelines (CPG) are an important instrument that translates knowledge and data into recommendations and unlike systematic reviews, commentaries and editorials, they provide conclusive recommendations to be used in action. The development of such guidelines is a scientifically rigorous task that requires subjective judgments. However, since these judgments are made by humans, they leave room for error and bias. Within the eurocancercoms project the issue of bias in CPG integration was examined by in depth policy analysis.
The sources of bias for the authors of CPGs vary and according to an article written by Allan S. Detsky, they can originate from the participants previously established experience or personal practice, which can dominate their views of evidence. Likewise, they can also originate from a possible long service to government committees or private insurance companies that could designate a different agenda on the experts minds, like the limited cost or other perspectives.
Nevertheless, the most commonly discussed form of bias and conflict of interest is financial. Such bias is usually noticed when there are ties between pharmaceutical companies and authors of CPGs. There are at least two common ways through which industry relationships can affect the content of CPGs. The first one is directly through a financial relationship between the guideline committee members and the representatives of the industry.
A recent study, aiming to quantify the extent and nature of interactions between authors of CPGs and pharmaceutical industry reported in the Journal of the American Medical Association, revealed that the 87% of the authors of CPGs had some form of interaction with pharmaceutical industry. The study highlights the need for appropriate disclosure of the financial conflicts of interest of the authors.
Another study aiming to examine the degree and type of financial ties between the authors of CPGs and the pharmaceutical industry, has concluded that ninety percent of the authors of 3 major CPGs in psychiatry had financial associations to drug manufacture companies, the products of which were explicitly or implicitly detected in the guidelines as recommended therapies. In addition, none of the financial relationships of the authors were disclosed in the CPG.
Choudhry and colleagues have documented that authors of CPGs published in the 1990s had many financial relations with pharmaceutical companies, the products of which were being discussed in the guidelines. A survey of those authors, reported that only the 7% of these authors believed that those relations influenced their opinions when the 19% of them believed that their coauthors were actually influenced.
Finally, a review, the purpose of which was to explore the extend and scope of American Academy of Neurology (AAN) guideline author, reported conflict of interests and implications for management. The report states that 46 of the 50 CPGs which were examined had at least one author with a conflict of interest. More precisely, the authors of the Quality Standards Subcommittee guidelines were more likely to have personal income conflict of interests with pharmaceutical and medical device companies whereas authors of Therapeutics and Technology Assessment guidelines were more likely to have clinical practicerelated COIs.
The second way, by which pharmaceutical industry influences the authors of clinical practice recommendations, is indirectly and lies in the process by which raw data becomes medical knowledge. Since the development of CPGs relies heavily on published evidence, there is a big chance that any industry involvement in the conduct of clinical trials and medical research publications in respected medical journals, can exercise significant influence on the authors perspective of view.
Recent literature discussing the influence of industry on medicine, reports that academic institutes are not left unaffected by the same influence, even though they are frequently regarded as free speech bastions of learning. That is also the conclusion of two former principal editors of the New England Journal of Medicine (NEJM), who state that it is almost impossible to find an expert who is not receiving payments from one or more drug companies in the filed.
Further concerns have also been expressed that pharmaceutical companies are involved in the field of research since they are capable of influencing study results. In addition there is evidence data showing that resulting outcomes from drug studies, which have been funded by pharmaceutical industries, are favorable to the funders.
Finally the integrity of medical reports seems to raise concerns as well. A recent study has revealed that 96% of the authors who wrote articles certifying the value of specific cardiac medications had a financial tie with the manufacturer. Further reports state that an amount of scientific articles appearing in widely known medical journals under the authorship of scientists and doctors, were actually written by ghostwriters on the payroll of industry.
Bibliography, reports several common forms (types) of financial influence that have been noticed in the process of developing CPGs. Pharmaceutical industry Sponsor, or other funding agencies can exercise influence in the selection or the guidelines topic, the selection of the expert panel members, the selection of technical consultants, the writing of the guidelines or even the final guideline document.
The role of social media in cancer communication
The relevance of social media to the eCancer domain results primarily from the fact that they can change (meaningfully) the way in which networking in the field is shaped, organized and maintained. A pre-requisite is to design tools, which will make users realize such benefits and internalize the new means for cooperation. Today this is a prominent challenge for professional communities, as it deviates from the way in which traditional practices are enacted and performed, while requiring a new culture for collaboration. The increasingly social conduct of work and the spin-off effects it bares on other relevant communities (i.e. neighboring disciplines, end users, etc) is clearly a motive. Nevertheless, appropriating the benefit is not straightforward.
For the cancer community the challenge is not setting yet another social networking site. Instead, it should aim to challenge the more demanding issue of crossing boundaries (or fields of expertise) and making new grounds for turning experience to collective know-how and know-why. Social media, designed suitably to foster cross-functional virtual communities and their practices in selected domains of eCancer, provide the means to this end. For the purposes of the present discussion and following the recent meeting in Crete, candidate areas for experimenting with boundary-spanning virtual practicing include:
Open access publication management
Cancer research management trials
Building evidence-based guidelines
Understanding the patients view and experience of care
Personal health records
Spanning boundaries Consider a professional research group specializing in cancer research. The group has just initiated a trial using a new drug, which seemed to work in laboratory settings. Therefore, they now wish to do the following: (a) disseminate their laboratory results to the wider community and (b) register their new trial.
Open access publication Attaining objective 1, the materials to be shared include a document, figures and images (linked to the document), sample visualizations (extracts of what was found in the lab), video, etc. The group leader enters ECCs and realizes that the available mechanisms for disseminating work include: (a) a paper following a specific template (i.e. keywords, abstract, single/double column, etc); (b) one-page leaflet following again a specified template; and (c) TV broadcasts. Choosing the preferred dissemination medium allows the group leader to populate an object (by uploading material, inserting figures, tagging, etc) and assigning some distinct services that the group would like readers to use so as to provide feedback. At the end of the day the groups work is updated and made available through a social networking site which compiles the collective wisdom on cancer research trials.
As shown, the Figure is not aimed to present merely the published material as accepted and uploaded at one point time. Instead, it emphasizes a living collection of relevant objects all contributing to gaining insight to how the initial contribution evolves with the input of third parties. As an example this presents this evolution in the fifth revision of the initial object collection. The activated links indicate the evidence added in this firth revision and how such evidence is classified (i.e. videos and pdf files).
It is important to note that the specific layout is only indicative, while we have intentionally chosen to organize the material in the manner depicted in web analysis. This organization reflects a classification of contributions by type of object (i.e. video, pdf file, figures) and keyword (i.e. drug) so as to enable further exploration within these categories. Thus, it is possible to query the available data by simply selecting interaction nodes scattered throughout the visualization. Finally, our choice of layout strategy is such that it allows us to establish links with the next scenario (see section 0), spanning the boundary between open access publication and cancer research trial management.
Managing cancer research trials Attaining objective 2, the information to be shared includes the type of trial, drug involved, treatment as a series of interventions, trial status / phase, sponsor, etc. Again, the group leader goes through some forms to do the task or in fact he can instruct the system to upload this information from another source (following a suitable schema).
The distinct feature of the visualization is that it classifies all collected experience in a manner suitable for quick overviews as well as visual exploration. Quick overviews allow users to gain insights to various data classification schemes. By presenting information on drugs (lower part) being investigated by the available cancer research trials, per phase (middle part presented as timeline) and the patient population split in social zones by gender (top part). The same data may be further explored by requesting more details on patients e.g. by revealing additional social zones by age group.
Further intuitive exploration of the data set may be facilitated by simple visual querying mechanism to reveal additional detail.
Extensions, relative benefits and value The above scenarios are only illustrative of what social networking can bring to eCancer practices such as research trials and open access publications. Moreover, the design concepts presented serve only the purpose of illustrating how tagging and other Web 2.0 technologies could be explored to this end. Obviously, as in other social networking applications the actual value is to be assessed only through time and by allowing users (of various types) to become engaged and provide contributions. Therefore, one could imagine various extensions of the scenarios sketched above so as to allow:
- Authors to revisit and change their data (without loosing earlier versions)
- Third parties to dispute basic conclusions in a study or
- Third parties doing similar things to inform the community of new sample data sets and invite commenting
- Patients to pose a relevant question
- Funding bodies to specify areas on research relevant to the drug
- etc
Living guidelines Reinterpreting the scenario presented in section 0 we could easily imagine a similar approach being applied and validated for guidelines. This time the documented material and its supporting evidence should be organized to reveal the content of a guideline and how it evolves in the course of time as new evidence (for and/or against) is accumulated.
Patient communities Patient communities can also benefit from a more informative exploitation of data as discussed in section 0. In what follows we present (hypothetical) examples of patients access to such repositories and the added value resulting from extracting information that reflects a personal profile or a point of view. It is important to note that our emphasis is not on typical community support functions (i.e. posting requests, receiving comments, sharing perspectives, etc), which are more or less available in social networking sites on health. Instead, we seek to understand how eCancer practices can be translated into meaningful and useful information in the users language. To this effect, advanced querying techniques can provide an important mechanism for disseminating research outcomes to the public as well as for classifying and presenting search results in a profile-relevant manner. Once gain our elaboration is purely exploratory and seeks to convey the concept rather than describe a particular solution.
Search = dynamic querying + exploration
Searching can take advantage from appropriate visual representations to present results in more intuitive manners. Let us imagine a patient interested in finding information broadly relevant to a medical profile. With current search engines, the way to approach the task is through keyword search. However, patients would appreciate presentation of search results in clustered collections or social zones of preference. Assuming some sort of mechanism / profiling technique whereby patients codify a social zone of preference, this presents the results of a Google search (the example depicts the search condition cancer research trials) in two different ways. The visualization presents the results by creating as many zones as the chosen classification scheme (i.e. number of different types of results such as web, video, images, books, etc.). Each result is presented as a node within a social zone and the size of the node is determined by the degree of interest, which in this case is the rank of the result within the data set.
Alternatively the same data are presented using a data mountain layout (i.e. nodes in the back of the graph get smaller indicating distance from the point of the users focus). This allows users to define the degree of interest by direct manipulation of the desire node. Furthermore, it would be possible to imagine scenarios of re-calculating the layout by changing / modifying the perspective of relevance.
Steps towards the solution
We have described preliminary results indicating how social media and Web2.0 technologies could be used in the context of EuroCancerComs to promote social networking within eCancer communities. The key idea behind the concepts is that users incrementally become content providers either by reflecting upon available knowledge or by establishing networking patterns. A critical technology is tagging which enhances the way of looking at a document, a web page, a speech, etc. Combining tagging with advanced and dynamic querying mechanisms is one way of engaging a variety of users in meaningful and recurrent exchanges in virtual space.
In light of the above, our vision of the ecancerHub social networking system can be summarized as follows:
Ultimate target: Design a virtual space to allow social networking on selected scenarios relevant to eCancer.
Context of operation: Social web site independent of existing portals (much like Google or Facebook) explicitly tasked to facilitate an emergent knowledge management process on eCancer digital content (various types and formats)
Properties of the social web: These may include:
Visualizing collective intelligence on eCancer
Replace currently prevailing form-based search with visual and exploration-based querying techniques
Interoperability with other systems on the grounds of widely accepted protocols and web services
To build such a system, we need to agree on the following:
Relevant information product families
What exactly is to be codified and how
Data types and formats to be supported
Indexing schemes, classifications, tagging
Flexible displays (plastic and role-based)
Orientation
Navigation
Exploration-based querying
Direct manipulation
Conjunctive versus disjunctive queries
Social aspects
Comments / annotations / voting / ranking
Negotiation / argumentation
Digital histories
The above could be tackled in a technical workshop aiming to address:
Community management
User roles, registration and access to information
Electronic neighborhoods (i.e. patients, professionals, mass media, funding bodies, etc)
Communication / notification services (threaded discussions moderation, asynchronous notification services)
Information content and interoperation with existing data bases and portals (e.g. dedicated portlets may be needed to assemble data from distributed sources and present them in a concise way; these portlets may be different if the material to be assembled is a paper and its resources from other types of medical data)
Examination of a suitable enterprise portal to serve as the community support system
eCancer practice
what practices are to be supported (i.e. cancer research trials, open access publication, etc)
for each practice we will need to define
- who is entitled to engage in a designated practice i.e. cancer research trials
- what processes are to be supported (i.e. registering the trial, reporting trials, conducting trials)
- indexing schemes and classifications (i.e. type of trial, drug, treatment / intervention, trial status / phase, sponsor, etc)
etc
Who does what, when and how (in a reasonable time schedule
TECHNOLOGY / ARCHITECURE
Text mining: developing and applying novel methodologies to cancer science information
The scienti?c literature provides descriptions of biological discoveries characterizing cancer at different levels of granularity. At the molecular level, articles contain detailed biochemical characterizations of genes and proteins in terms of sequence, structural and functional features. To detect associations between particular genes and diseases or phenotypic properties related to cancer, considerable amounts of population, epidemiologic, or even gene knockout studies using animal models have been published. The literature plays a central role in the current scienti?c discovery process, from the initial stage of experiment planning to the ?nal step of result interpretation and the subsequent generation of new hypotheses. Interpretation of genetic alterations observed in cancer patients using information hidden in the vast volume of scientific articles is increasingly challenging.
Eurocancercoms was able to map, describe and integrate the state of the art in the application of text-mining techniques to understand the cancer disease. It also described two novel text-mining methodologies developed and implemented by the consortium to discover genes that may directly implicated in malignant melanoma, and which are flexible enough to be adapted to other cancer types.
Based on the melanomas section of the National Cancer Institute (NCI) website a query to search PubMed was built. The query combined several MESH terms: melanoma[majr] AND human [mh] AND english[la]) OR (melanoma*[ti]). We restricted the search to articles published in the last ten years. As a result we obtained a list of 15,525 melanoma articles.
The second step filtered out those articles that do not contain molecular information in their abstracts. Many of the articles collected in the first step included topics like surgery, treatments guidelines, etc. that doesnt bring any light about the genes that are possible involved in this disease. The ABNER software was used to tag the bio-entities (protein, DNA and RNA mentions) in the abstracts automatically, retaining only those that contained mentions of these bioentities. After the filtering process, 4,580 articles tagged with their corresponding genes and proteins were recovered. The resulting set constituted the positive training data for the subsequent semi-supervised machine learning based document classifier. In order to define a negative (non-relevant) training set, a random list of 4580 papers (balanced data set) containing genes or proteins in their abstracts but not directly associated to melanoma was created.
Pipeline that summarizes the novel method developed by eurocancercoms consortium to rank the genes according to their association with melanoma. To classify the documents we used a Support Vector Machine (SVM) with a linear kernel function using the SVMLight and LibSVM packages. The classifier was trained to generated a predictive model and validated through five cross-validation.
In the second approach a list of melanoma genes extracted from the Biocards section of the MMMP database was extracted. The goal in this case was to find a possible functional interpretation of the genes based on literature analysis. SENT was used which is a functional interpretation tool based on literature analysis. SENT uses Non-negative Matrix Factorization (NMF) to identify topics in the scientific articles related to a collection of genes or their products, and uses them to group and summarize these genes.
SENT clustered the genes in 8 groups, for each group the literature terms, the genes names and the number of articles supporting the decision where displayed. These results provide a general idea of the topics latent in the literature for the MMMP genes.
The results of the two strategies followed by the eurocancercoms consortium to utilise text-mining methods to understand the relation between genes and the melanoma disease provided the first novel scientific approach to integrating science communication to enhance dissemination and utility of the published literature thus providing the first proof of principle aimed at addressing info-overload.
Architectural approaches to cancer science information: analysis & integration
Eurocancercoms has been continuously working on an application that allows the research community to explore the cancer disease from an integrated point of view. Our aim was to take advance of the revolution in the analysis of genetic diversity in humans and its role in disease that has come with the advent of massive parallel sequencing (MPS) strategies. These methods allow the acquisition of genetic information on a massive scale (at a cost of somewhat reduced accuracy) and low price.
New large-scale genomics efforts are generating data at unprecedented scale. The 1000 Genomes Project (www.1000genomes.org) alone is expected to produce 60 times more data than has been deposited in the public repositories over the last 25 years, and has already surpassed the amount of sequence data in Genbank. The information generated by projects such as The 1000 Genomes Project will allow us to find relevant variants for different cancer types.
In a first iteration somatic mutations were extracted from the COSMIC database and germ-line mutations from dbSNP. For each gene the application displays the number of germline and somatic mutations found, and details about each mutation.
In this new version we added information about pathways extracted from the KEGG database (http://www.genome.jp/kegg/). It has been proposed that the genes mutated in cancer tend to encode proteins that cluster in certain signalling pathways, and different sets of genes related to specific tumours might, in fact, correspond to a unified set of affected pathways or functional processes. This pathway-mapping approach has been applied in order to discover pathways that may be particularly relevant in melanoma.
Currently the application provides also Information about drugs that may be used to treat the disease. We collected the information from NCI (http://www.cancer.gov/) PharmaGKB (http://www.pharmgkb.org/) and Matador (http://matador.embl.de/). The relation between a drug and a gene may be a direct or indirect. In the first case the gene is a direct target for the drug. Indirect relations are established based on information derived from literature or from pathways analysis. We are currently working on categorizing the information about drugs to offer a more complete scenario to understand the drug activity.
Finally we also introduced the results generated by the two text-mining approaches described before, as a way to understand the possible implications of the genes in melanoma. For each gene we provide the rank obtained from the SVM classifier and also the terms obtained in the clustering analysis.
The application is available at: http://eurocancercoms.bioinfo.cnio.es. The main window contains a table with the MMMP gene list and the information about mutations, pathways, drugs, cancer types where the gene plays a relevant role and the literature score provided by the SVM classifier.
Proto-type for e-engagement for prevention (nutrition) and R&D
Nutritional status is an important factor in prediction of the risks associated with cancer disease and treatment affecting clinical outcomes, including survival and the ability to tolerate treatments. Nutrition management has been proposed as auspicious to cancer patients because the nutritional status and the nutritional intake are relevant factors affecting also the quality of life and the psychological status of cancer patients. Nutritional screening has been identified for early intervention in clinical practices, however there is not always proactive strategy that can be used when the patient is discharged to home. The number of cancer survivors continues to increase due to many factors, including a growing aging population, early detection, improved diagnostic methods, more effective treatment, and improved clinical follow-up after treatment. It is important for physicians, patients and public health professionals to be knowledgeable of issues survivors may face, especially effects of treatment on their physical and psychosocial well-being. This understanding is critical in promoting good health and coordinating comprehensive care for cancer survivors. Furthermore, a significant proportion of the worldwide burden of cancer could be prevented through the application of existing cancer control knowledge, implementing programs and public health campaigns promoting, for instance, physical activity and healthier dietary patterns.
The world wide web has changes the mass and health communication in the last decades. The Internet and the associated technologies can contribute to disease prevention and health promotion in a number of contexts, enabling a swift, inexpensive and easy distribution of content information; this may turn experience into learning. The Internet offers potential for interaction and is a medium for influencing learning and behaviour change. Evidences from randomized studies suggest support through Internet to cancer survivors in order to reach a positive outcome regarding social functioning, ability in searching for information, depression and state of health perceived though cancer patients quality of life seems not to have improved.
This S&T result was derived from a collaboration between eurocancercoms and non-eurcancerom partners: the Organisation of European Cancer Institutes (OECI), the Italian Association for cancer patients, their families and friends (AIMaC - Associazione Italiana dei Malati di Cancro) and the Italian Federation of Volunteer-Based Cancer Organizations (FAVO - Federazione italiana delle Associazioni di Volontariato di Oncologia).
The change in quality of life, psychological status and nutritional knowledge, measured by study questionnaires was used to evaluate the effectiveness of the prevention approach based on the 6-month visits of a website dedicated to study. This novel architecture - Internet space where tItalian patients could find information about nutritional care and where to interact (forum and blog), was shown to positively improve knowledge, quality of life and psychological status of the oncological subjects.
Integrating cancer communication & information: a demonstrator social media platform ecancerHub
Opus: Core platform
Opus is Reality Digitals tried and tested video and community and social media platform and comes as standard pre-configured with the following functionality:
Secure registration
Secure login
Single sign on
Profile creation
Blogs
Forums
Friends
Groups
Google Map integration
Video ,Audio, Graphic and Text upload via:
- Web
- Webcam
- Mobile
- Camera
- Bulk upload
Content syndication too: Youtube/Facebook/Micro-site (if required)
Comments
Ratings
Industry leading moderation dashboard
Extensive administration and auditing dashboard
eCancer Community functionality
Homepage
Header Displayed on all pages
Logo
Lead site sponsor logo
Leader board advertising banner
Navigation bar (Standard Opus functionality) displayed on all pages
Home
Articles
People (members)
Groups
Videos
Blogs
Chat and Answers
My eCancer Hub (only viewable after member login) Prompt to login
Site search
Homepage content sections: Editorial (left hand iFrame)
Content box Welcome copy
Content box Latest on eCancer Hub (list of editorial topics from publishing partner)
Content updated automatically via RSS or XML
Content box Latest Chat topics
Automatic Pull from chat section - Opus
Content box Video wall (Member and Professional content)
Automatic Pull from video section
Content box - Featured members
Automatic pull from member profile section
Permanent commercial content displayed on all pages (right hand iFrame)
Content box Login or register section
MPU and Sky Scrapper ad spaces
Market research polling functionality
Content box Useful links section
Content box Sponsored section
Content box Latest News
Content box Site sponsors
Registration Page Opening in left hand iFrame
Blank box User name
Blank box Name
Blank box Surname
Blank box Password
Blank box - Email address
Drop down Country
Blank box Postcode
Drop down User group
Large blank text box Tell us about yourself
Check box Send weekly news letter
Button Enter
Email notification of accepted membership, auto generated and sent to new member
As a member youll also be able to:
Write or read blogs
Ask or answer queries in our chat forum or have a good old moan!
Comment on any articles you read
Post your videos into our galleries
Meet a huge network of cancer professionals
Get advice
To bookmark ECancer Hub (so you never have to search for us) click here.
Profile page Accessed though My ECancer Hub button on the Nav bar.
Profile tab
As registration page functionality +
Upload picture
Save and edit button, so you can edit Tell us about yourself text box
Navigation tabs on top of Profile page (next too Profile tab)
My Friends
Displays current friends and pending approval friends
Displays total number of friends
Button Find new friends
Find new friends opens the People/member page
Click on a member and view their profile
Button Invite as friend
Opens text box write a message to person inviting them to be a friend
Button Send invite
My Groups
Displays current subscribed groups and group invitations
Button Create a group
Opens Create a Group page
Text box - Group name
Text box Describe your group
Drop down Select your subject
Check box Private
Check box Public
Button - Create
My photos and videos
Displays uploaded photos and videos
Button Add new photos or videos
Opens upload media page
Text box Browse for media (on your PC)
Text box Name your photo or video
Text box Date taken
Check boxes Who should be able to view this
Public, Private, Friends
Drop down Publish to gallery
Text box Tell us all about it
Button Upload
Button Cancel
Button Add another
Articles Page- Accessed through either the Top navigation bar or by clicking on an article in the Latest from ECancer Hub section on the homepage
Opens into the left hand iFrame on the page.
Lists out the latest stories/articles on eCancer Hub
Picture graphic and brief copy displayed for each story or article
Content is auto feed from publisher via RSS or XML
Click on story/article
Full story or article opens into the left hand iFrame
Potential hyper links within the story highlighted in blue
Hyper links open into left hand iFrame
Articles can be sorted by: Most recent or most read
Button Back (cycles you back to main articles page)
Reporting
ECancer Hub will be provided with 24x7 access into all of Reality Digitals Opus reporting tools.
ECancer Hub will be able to monitor all site activity and capture critical metrics in real time.
ECancer Hub will be provided administration tools to QA clips/pictures, comments and ratings before publishing: Ensures quality and content standards are met before publishing
Omniture or Google tags are easily added for additional reporting depth (upon request)
Reporting Metrics Include:
File List - Detailed list of all media files on the site
User List - Detailed list of all users and registration information
Group List - Detailed list of all groups and activity
Top List - List of the top 100 media files by date, number of views, ratings or votes
Daily Report- Daily activity report of number of users, registrations, uploads, media views, comments, ratings, etc.
Encoder Status - Real-time status of all clips being encoded
Web Logs - Reports on server activity/traffic
Bandwidth Report - Daily and monthly reports on bandwidth usage
Moderation Tools:
The site administrators have access to a robust set of moderation tools via the Opus Dash¬board, which allows for monitoring and reviewing of User submissions before pub¬lishing to the site and appropriate actions. Moderation tools include:
Pending List - Detailed list of all media awaiting approval with approve, re¬ject, and delete functions
Blog Admin - Administer all blogs on the site
Comment Report - Detailed list of all comments and approval functions
Blocked IPs - Configuration of IP addresses banned from the site
At the user-level, users have the ability to flag content as inappropriate.
Filtering
The administrator can identify a list of offensive words or phrases that prompt fur¬ther action or automatic deletion. (English only unless customer provides international word list)
Video Branding
ECancer Hub videos can be branded with pre-roll and post-roll video which stays with the video wherever it is distributed.
SEO
Search engine optimization (SEO) of content will be achieved through HTML meta-tags and alternate content
Hosting Support and CDN
The entire ECancer Hub program will be hosted and supported by Reality Digital. SAN storage systems are monitored 24x7 and are fully redundant and backed up hourly to NAS. Daily tapes and backed up and removed from site.
Reality Digital provides system level support across the entire solution including Apps, backend and CDN. Customers are provided with all relevant contact information.
For DHP we will leverage the Akamai CDN to dramatically improve User experience from most anywhere in the UK and for than matter on the planet.
Storage & Search
Reality Digital Opus is based on industry standards such as Microsoft COM and ASP, SOAP, XML, HTTP and SMTP. The content in the database can be easily exported into an XML without impacting any existing systems.
Storage is architected on a redundant fiber connected SANs replicated to a NAS for additional redundancy. Additionally, Flash media is populated on the CDN at the time of ingest and encode.
Reality Digital offers SQL based search. The users can search the content using keywords and metadata and a user by the username, name or email address. Optionally, we can integrate with any 3rd party Search Engine to support more advanced search features. We have experience with most major search applications. The addition of one of these search engines adds significant expense, but all rich capabilities for search, recommendation, and highly customized user experiences.
Data Protection
1. Reality Digital powers all sites on redundant database servers to ensure availability
2. RAID 5 SANs continually replicated to NAS devices for back-up of all data and media
3. Nightly backup of servers
4. Nightly backup of data to tape Weekly rotation of tape is stored offsite
5. Virtual clones of web servers allow us to bring up a replacement within minutes
6. In some cases, where load balancing is part of the configuration, we have several identically configured servers standing by to prevent down time
7. Redundant Switches with redundant uplinks ensure constant connectivity to the internet
8. Redundant Encoding and Ingest services provide fast availability of incoming media
9. Limelight and Akamai CDN integration provides additional data redundancy of content
Third Party Integration: integration with many third party technologies including: Adobe, Akamai, Doubleclick, Google Maps, Limelight, Microsoft, Omniture, Paypal, Word¬press, and many more, including: eCommerce and retail customers.
Additionally through the template development work with TME, integration of the event registration and booking systems should be straight forward.
Professional Services and Program Delivery
Because the Reality Digital platform has already been built out, the actual development team required is very small. A typical project requires a project manager, a project architect, a designer (as required) and a QA team. Some projects demand a second engineer and a Flash developer as well. Because we build social video sites day in and day out, a typical project will take from 6 8 weeks all in to deliver.
Potential Impact:
The project has been distributed into the public domain through two urls:
1. www.eurocancercoms.eu/
2. http://www.ecancerhub.eu/
In addition there have been numerous non-publication e-distribution resources. For example,
Engaging the patient community: round-table and media distribution
Engagement using You-Tube
Extensive e-banner and distributed web-site advertising and PR
ecancerHub: a platform for developing novel cancer science information tools
Impact and dissemination are listed in following pages.
List of Websites:
1. www.eurocancercoms.eu/
2. http://www.ecancerhub.eu/