partially”
Relevant data to answer the research question may exist elsewhere, for instance in
clinical laboratories providing diagnostic tests. A registry for the same disease may
already be in place in another country. In some situations, adding new data to an existing
registry rather than setting up an entirely new registry may be more cost-effective.
Orphanet maintains a database of PR in the field of RD which can be consulted online. In
any case, consideration should also be given to the interoperability of registers in the
same field, to allow exchange of data and availability of a common subset of data for
common analysis.
“Deciding whether the registry is the most appropriate instrument for addressing the
research question”
This is crucial as time and resources needed to collect and process data in a registry
setting can be substantial. This question should be considered in partnership with
epidemiologists who are familiar with all possible data collection types with their
respective advantages and disadvantages. For instance if the primary goal is to establish
the incidence and the prevalence of a rare disease, a possible approach is to perform a
“capture-recapture study” using all possible sources of data to “capture” cases in a given
geographical area. For this goal, it is not necessary to have a permanent registration. As a
principle, PR should only be created when there is a need to collect information over a
long term. When the data collection requirements are temporary, they are more
efficiently handled through a research study. Once a registry is created with an
administrative structure and staff, it may be difficult to terminate it even if it has served
its purpose.
“Identifying the stakeholders to envisage a partnership with them”
Other stakeholders are likely to have their own interest in accessing good quality data
and may be willing to share the workload and the cost. The stakeholders to
systematically consider are:
(1)
the regulatory authorities, the health care providers and the payers and
commissioning authorities who will wish to access data to assess the
impact of interventions, especially of new drugs, new protocols;
(2)
the product manufacturers who often get a conditional market
authorisation for their orphan product which obliges them to establish a
post marketing follow-up, usually through a registry. Manufacturers are
RDTF Report on patient registries in the field of rare diseases (June 2011)
11
also interested in accessing data on the natural history of a disease for
which they have a product in development to assess the needs and the
feasibility of the development;
(3)
the treating physician groups, the universities or professional societies and
researchers in charge of clinical research;
(4)
patient organisations which rightly consider that gathering data on their
disease is pivotal in motivating academic researchers and Industry to
invest on their disease. Patient groups spend considerable amounts of
core resources on data collection and are invaluable partners at all stages
of establishing and running a registry.
“Defining the scope of the registry”
This includes the planned representativeness of the target population and the
characteristics of the data to be collected. When the purpose of the registry is defined,
the next step is to define the data to be collected to reach the define goal(s). The scope
of the registry is defined in terms of size (targeted number of patients), setting,
geographical coverage and anticipated duration. The scope is adapted to the ambition of
the promoters with due consideration for feasibility. The data to be collected is defined
jointly with the final users of the data, as the format of the data pre-empts its potential
utility for research purposes. If the data are not appropriate or are not formatted
appropriately, it will be impossible to make sense of the data and use them.
“Assessing the feasibility of the project and the likelihood that it will be a success”
Establishing a registry is an investment which requires a careful approach. A feasibility
study should take into account the following elements: Commitment of all professionals
at the origin of the data; accessibility of the data (physically and legally); commitment of
professionals to be involved in the monitoring and the analysis of the data; anticipated
availability of an appropriate tool to collect, store and analyse the data; outreach
programme to educate and promote adoption and proper use; possible funding for the
initial phase; and plans for funding long-term activities.
“Identifying possible sources of funding, including long-term funding”
Funding should be envisaged not only to develop the tool to collect data, as often seen,
but also to monitor their quality and to analyse them and disseminate the results.
Usually the budget has to be substantial and is often underestimated at the beginning,
when the promoters are not used to the epidemiology field. In many cases it takes years
to realise the full benefits of a PR. Therefore long term funding is crucial. It is a waste of
resources to go through the developmental phase of a registry only to have it terminated
because of lack of funding.
Sources of funding are different from one country to another, usually similar to
sources of funding for research activities, because most registries in the field of rare
diseases are built up to serve an identified research goal. In that case, funding is likely to
be limited in time, leading to a crisis at the end of the grant.
Countries may have a specific mechanism to fund registries, usually cancer registries,
birth defects registries or cardio-vascular disease registries, but the committees in
charge of assessing proposals are not sensitised to the specificities of patient registries in
the field of rare diseases.
RDTF Report on patient registries in the field of rare diseases (June 2011)
12
Funding from the European Commission is accorded on the same grounds as national
grants. Both DG Research and DG Sanco provide funding for PR in the field of RD (for DG
Research, funding is only granted is the PR is a tool for research purposes), but for
limited time period although the European Commission acknowledges the need for a
more sustainable support.
Long-term funding challenges for rare disease PR need to be addressed by pooling
non-profit and for-profit resources in the most effective way.
As a consequence a reasonable funding plan is to devise each stage of funding as
having its own results, and not being just an intermediary step to a long term result or
even "hope" (e.g. availability of patients for research, without knowing precisely what
research). Even the pilot phase can if at all possible tackle a question of relevance, so
that pilot funding serves a greater purpose than simply establishing feasibility. Each
funding application needs to be able to demonstrate the feasibility of reaching one or
more objectives within the funding period, regardless of whether funding thereafter
continues. Funding should where possible be used to enhance sustainability for the
future (e.g. the development of software may involve short terms costs, but reduce long
term costs) but not completely at the expense of producing any short term useful
results.
Steps to consider when developing a PR in the field of
RD
Establishing a registry is a complex process which requires a range of technical and
organisational skills for the process to result in an effective data system. There are eight
requirements that are critical to the successful development of a new PR.
Implementation Plan
There should be an implementation plan starting with a pilot phase when procedures are
carefully evaluated and refined. It is more efficient to identify and resolve problems during a
pilot phase than to invest a great deal of effort and resources developing software and
procedures and acquiring equipments that must be altered later. In order to maximise
adoption the application must be easy to use. It is important to build self-explanatory web
forms tested with the people who will use them be they patient or clinician.
Adequate registry documentation
Adequate documentation is essential for ensuring the quality and efficient operation of the
registry. It should include a definition of who will operate the PR; a thorough description of
the inclusion and exclusion criteria; definitions of data sources, data collection, data editing
and data entry procedures; data processing procedures; hardware and software manuals;
definition of analyses that will be routinely conducted; confidentiality guidelines and access
rules.
RDTF Report on patient registries in the field of rare diseases (June 2011)
13
Data quality
The importance of data quality and data completeness assurance measures should be
highlighted as pre-requisite for ensuring meaningful analyses and outcomes. Registries are
(usually) voluntary programs, and the data quality may depend on the motivation or
commitment of health care professionals submitting the data.
In addition, test results may not be comparable due to differences in test methods and
equipment, as well as some methods of evaluation which are not-standardised between
institutions. These may influence the analyses of the data and hence the outcomes.
In view of this, the realistic expectations from the database should be established, and
the careful quality monitoring and quality improvement, as well as motivation of the health
care professionals performing the data entry, have to be implemented. In addition, the
voluntary and observational nature of registries necessitate that analyses and reports be
undertaken with the appropriate consideration of potential cohort biases, confounders and
incomplete data sets, using standard epidemiologic methodology.
Flexibility
Given the constant progress in scientific research, ensuring “flexibility” of the data base for
research questions is critical to be able to respond to the research needs. For example, the
minimum data elements may need to be subject to regular change and adaptation.
Technology
The data base has to be user-friendly and easy-to-use. The newer technologies are better,
but usually more expensive. It is important to note that the change in a technology (for
example implementation of a newer and a better software platform to capture the data)
might have (considerable) financial and staffing implications.
Determination of data elements
One should be realistic when defining the data to be collected, bearing in mind the resources
available and likely compliance of collectors as well as potential value for research. The
definition of data elements should conform to professional standards and to International
nomenclature when they exist. Coding in non standard format makes the merging of
information with other PR virtually impossible.
Ways to pool data and connect databases
Interconnecting data bases may have some limitations which need to be taken into
consideration, and of which the most important are:
1.
Legal considerations relating to data privacy and data protection. For
example, for processing of data for a combined analysis, in order to export
the data from 1 database, new protocols (including these combined
analysis) have to be established, and all enrolled patients have to be re-
consented;
RDTF Report on patient registries in the field of rare diseases (June 2011)
14
2.
Technological barriers, such as differences in software, may preclude
processing of data for a combined analysis, or limit possibilities for
comparisons.
The type of software to be selected should be adapted to the specificities of the diseases and
of the data collection organisation. The possibilities include:
(1) An off line system:
A unique central database stores all the data.
This is a very secure and inexpensive system. When collected at distance, data can be
sent by post, by fax, by e-mail, although these systems have their limitations in terms
of security. This is a less costly approach as non commercial soft wares for managing a
unique database are available and can be easily customised after a short training.
As many local databases as there are entry points for data collection store
their own data.
The data is transferred from time to time to a central registry (EUROCAT
model). This model is a bit more costly as there is a need for software at each point of
data collection, but is still a relatively low cost approach.
(2)
An online system:
All data collectors have the right to electronically submit data to the distant
database either by physicians or patients.
This solution can require high costs in terms of initial development and maintenance
of the online system, raising many legal and security issues when the data are
collected in one country and stored in another.
Both models have pros and cons depending on the data to be collected, the number of
places where data are collected in one or more countries, and the choices of the
different partners.
Whatever system is used, it should be possible for the provider of the data to be able to
see easily what data they have provided, correct it if necessary, and produce local data
reports.
It is recommended that Member States and/or the European Commission facilitate a
centralised mainframe for data collection, storage and analysis, to allow long term
maintenance and reduced costs.
Data access policy
The ability to access information in an efficient, flexible and timely fashion is a key element to
the success of a PR. However PR store highly personal and sensitive information. Therefore
the confidentiality of the information must be guaranteed. Allowing access to PR data
without breaching confidentiality requires thought and planning. Establishing a policy for
data access is a crucial task to guarantee the protection of confidential data and maximise
the use of data by all stakeholders.
RDTF Report on patient registries in the field of rare diseases (June 2011)
15
Framework for disseminating data and findings
It is necessary to define the routine data products and reports that will be needed to fulfil the
goals of the registry. In addition the data should be used as much as possible to maximise
output. The value of a PR is limited by the extent to which the data are analysed and results
disseminated to relevant audiences. Nowadays the availability of data products and reports
on a website is required.
Part II
Ethical and legal issues
Serious ethical concerns have led to legal requirements for using health information for
research purposes. The creation and use of PR for research purpose constitute “research
involving human subjects”. Legal requirements exist on regional, national and European level,
but are not always the same. Many PR have difficulties consolidating this information and
fulfilling the various obligations. These legal requirements may influence the selection of data
elements and the way verification can be organised, as well as affecting the subsequent use
of data for secondary research purposes.
The purpose of the PR, the status of its developer and the extent to which registry
data are identifiable largely determine applicable regulatory requirements.
Ethical concerns about the conduct of biomedical research and the use of health
information in the past has led to a call for the establishment of a code of conduct for the
various third parties who usually have a contact in this process: the patient organisations, the
academic researchers, the policy makers and the Industry.
Consent requirement
The ethical principle of respect for persons supports the practice of obtaining individuals’
consent to the use of their health information for research purposes. This includes consent to
registry creation by the compilation of patient data, consent to the initial research purpose
and uses of registry data, and consent to subsequent use of data by the PR developer or
others, for the same purpose or other research purposes.
Individuals (children and parents included) should be informed with respect to the type of
research that will or might be carried out, the arrangements for access to or sharing of stored
information, and the duration of storage.
The consent process should also include instructions about the way to withdraw at any time.
Conflicts of interest may result in undue influence on patients and compromise voluntary
RDTF Report on patient registries in the field of rare diseases (June 2011)
16
participation. Consent should be given freely, free from pressure or persuasion, based on
information provided by trained staff. If consent is withdrawn, the data is no longer usable.
When data is collected for research use, an ethics review committee will oversee the process
and will be required to assess and ensure that benefits of the goals of research outweigh the
risks of participation.
Information and consent should be obtained in writing and specific protections should be
provided for vulnerable subjects and vulnerable populations, based on the general principle
of acting in their best interest.
If the data collection is undertaken by a group from a different country, regulations from
both the country of origin of the data and of the country where the data is stored should be
respected in order to maximise the protection of the rights of the investigated group of
patients.
The applicable regulations are different depending on the model of registration chosen:
The first model is to register contact data, demographic and diagnosis data or
exposure data, and to re-contact the registered individuals every time that a research
study is initiated, to invite them to participate. This is the best model in terms of
respect for persons, as individuals will have the opportunity to participate or not in
any new study. This requires keeping track of people’s movements or of patient
deaths: most legal frameworks require that all persons registered revalidate their
data annually.
The second model is to register all needed data for an identified research purpose
and to provide access to the dataset to a permanent, well-identified group of
researchers. No disclosures of data will occur and all research activities have the same
scientific purpose.
The third model is to register data for an initial specific purpose with the clear
intention to provide access to third parties and to subsequently use the data for other
not yet identified purposes. This is the most complicated model, requiring a lot of
attention in terms of protection of data privacy and respect for persons.
In the case of existing data collections, investigators should be required to re-contact
subjects to obtain consent for new studies. If it is impossible or impractical to gain consent,
an appropriate ethics review board must give its consent for the further use of the data.
Previously collected anonymous data, irreversibly anonymised, may be used for purposes
other than those originally intended.
The decision to strip datasets of identifiers irreversibly requires careful consideration. The
benefit of having unlinked anonymised data is to ensure absolute confidentiality thereby
allowing further use of the data. However, retaining identifiers, though requiring further
consent from the subject, permits more effective biomedical research and the possibility of
re-contacting the subject.
RDTF Report on patient registries in the field of rare diseases (June 2011)
Dostları ilə paylaş: |