Sherman, Philip M.; Brown, Sally; Rose, Keeley; Murphy, Tim; Makarchuk, Mary-Jo; Paterson, William; Belanger, Paul
Crohn disease and ulcerative colitis are chronic, debilitating conditions that affect the gastrointestinal tract (1). Although they are distinctly different illnesses, they share many clinical, epidemiological, and immunological characteristics and thus are commonly grouped together as idiopathic chronic inflammatory bowel diseases (IBDs). Approximately one-fourth of all patients with these disorders present during their childhood years. On the basis of epidemiological studies, it has been suggested that the incidence rate of IBDs in Canada is among the highest in the world (2). In Ontario, the incidence of pediatric IBD was 9.54 cases per 100,000 in 1994, rising to 11.4 cases per 100,000 in 2005 (3). These data indicate an increase in the incidence of pediatric IBD in Canada, as well as an increase in prevalence among children in certain ethnic communities. Pediatric IBD has many characteristics that distinguish it from adult-onset IBD (4), yet management of pediatric IBD has largely been based on studies in adults. Important clinical questions remain unanswered, and the factors that trigger these diseases remain, to a large extent, unclear.
The burden of these disorders on patients and their families is high. For children with IBD, especially Crohn disease, the onset of puberty, linear growth velocity, and skeletal development are often impaired (5). Moreover, children and adolescents with IBD have an impaired quality of life and relatively high rates of depression and anxiety (6), in part because of the waxing and waning course of the chronic illness.
In June 2011, the Foundation for Children with Intestinal and Liver Disorders (CHILD) and the Canadian Institutes of Health Research (CIHR) Institute of Nutrition, Metabolism and Diabetes (INMD) convened a workshop in Toronto, Canada, entitled Developing a Pediatric Inflammatory Bowel Diseases Network and Data Platform in Canada. A group of pediatric IBD researchers from across the country were invited to participate in the workshop with the objectives of identifying the scientific goals for a Canadian pediatric IBD network, the required infrastructure, and an appropriate governance structure needed to achieve a network and data platform, as well as other elements needed to ensure the successful implementation of a network.
A network is a system or set of connections between groups. In the present case, an IBD network is defined as a system of connected hospitals and clinics that treat pediatric IBD in Canada. A platform may be defined as an underlying computer system on which application programs can run. The data platform that we will develop is a computer system for storing individual patient records and relevant health information on which system administrators and health researchers can run programs to generate descriptive statistical information to study various aspects of IBD.
CHILD was established in 1995 to fund research for children with Crohn disease, ulcerative colitis, and liver disorders. Located in Vancouver, British Columbia, CHILD is committed to finding a cure for these disorders.
The CIHR is the Canadian government's agency responsible for funding health research. The INMD is 1 of 13 virtual institutes that comprise CIHR. The mandate of the INMD is to support research to enhance health in relation to diet, digestion, excretion, and metabolism and to address the causes, prevention, screening, diagnosis, treatment, support systems, and palliation for a wide range of conditions and problems associated with hormone, digestive system, kidney, and liver function. Relationships between CHILD, CIHR–INMD, and other organizations are provided in Table 1.
The context for the development of a national pediatric IBD network included consideration of the CIHR Strategy for Patient-Oriented Research and lessons learned from a number of other similar international and national initiatives. Announced in August 2011, Canada's Strategy for Patient-Oriented Research (SPOR) was developed by the CIHR in collaboration with provinces and territories, health charities, academic health care organizations, and industry representatives. The mission of SPOR is to improve health outcomes through clinical research by addressing gaps in the research-to-practice continuum. The scope of SPOR includes the continuum from ‘first-in-patient’ studies through to the integration of drugs, devices, and procedures into health systems. As the term patient-oriented research implies, SPOR proposes to create an active and engaged role for patients and their families in research.
The Canadian Maternal Infant and Child and Youth Research Network (MICYRN) arose from a coalition of child and maternal health researchers and institutions with the mandate to build capacity for high-quality clinical research in the area of child and maternal health in Canada. The vision of MICYRN is that the synergistic strength gained from a formal collaborative network of researchers enabled by integrated infrastructure will spur advances that are unattainable by individual research groups. Although the MICYRN network is different from the proposed pediatric IBD network, MICYRN is supportive of networking child and maternal health research and may be viewed as a potential prototype of a national disease-oriented network that can be showcased as an example for others to follow.
ANALYSIS OF PREVIOUS PEDIATRIC IBD NETWORKS
A national survey of pediatric gastroenterologists, conducted in Fall 2009 showed strong support for the concept of a national pediatric IBD network and data platform. The survey highlighted some of the critical success factors for the initiative, including the provision of sustainable, long-term financial support, and the requirement of having an identified scientific champion and leader to guide the development process. In addition, the survey highlighted several key challenges in the present research system, including the lack of coordination and duplication of technology, the lack of critical mass of patients and scientists in pediatric IBD in Canada, privacy concerns, and misaligned systemwide incentives.
CHILD subsequently held an international advisory workshop in June 2010 in Vancouver to learn from similar initiatives in other parts of Canada, the United States, and the United Kingdom and to solicit advice on the design and implementation of a national network in Canada. Seven themes emerged from the workshop (Table 2).
Two examples of previous unsuccessful attempts to create pediatric IBD registries in other jurisdictions highlighted the challenges encountered: a lack of clear direction from the leadership, limited resources available for submitting centers to enter patients into the registry, overload of limited local resources resulting in poor patient enrollment, noninclusivity (ie, the network included only a limited number of centers), lack of linkage between funding and patient recruitment, inadequate oversight, lack of predetermined scientific and clinical rationale to support network existence, and a leadership that was seen as not responsive to advice.
The Pediatric Resource Organization for Kids with Inflammatory Intestinal Disease (PRO-KIIDS) was established 3 years ago by the Crohn's and Colitis Foundation of America (CCFA) to set out research priorities and to further pediatric IBD treatment and research. The CCFA Pediatric Network is the first major pediatric IBD research initiative under PRO-KIIDS and involves various institutions working together on mutually agreed-upon projects. A combination of CCFA representatives, network investigators, and an external advisory board helps steer the network and monitor network activities moving forward. A risk stratification study is the first project under way and is fully funded by the CCFA. The risk stratification study has benefited significantly from the network through accelerated patient accrual as a result of standardized processes across network members and the concept of collaboration toward a common cause. An additional benefit is the cost-effective ability to add clinical sites to study design, when required.
Identifying the Parameters of a Canadian Pediatric IBD Network and Data Platform
Workshop participants were asked to work in small subgroups to identify the scientific goals, infrastructure requirements, and appropriate governance structure for a Canadian pediatric IBD network.
Objective 1: Defining the Scientific Goals and Advantages of a Canadian Pediatric IBD Network
Although no overall consensus on goals was reached, there was convergence on several themes, which are summarized in Table 3. Suggested scientific goals were to develop clinical and research-relevant measures to identify the risk of developing IBD; to assess the psychosocial affect of the disease on the child and family, including indicators of psychosocial risk and taking the opportunity to respond to patients’ concerns regarding quality of life issues; to track the epidemiology, the location, and the ethnicity of IBD, including evaluating particular populations and ethnicities (eg, Jewish, south Asian); to describe patient cytokine profiles and how they can be used to target therapies; to understand reasons for the observed increasing incidence of IBD in the pediatric population in Canada; to explore potential new uses for old drugs and determine the patient subgroups in which these drugs work best; and to define the role of epigenetics in pediatric IBD. Several participants favored a focus on very-early-onset cases of IBD and disease progression over time as a main driving theme.
Patients and their families often request that the Crohn's and Colitis Foundation of Canada (CCFC) supports research into the effects of therapy (eg, on future reproductive health) and the potential role of alternative medicines, complementary therapies, and nutrition in enhancing quality of life. Researchers and patient representatives agreed that the focus of a Canadian IBD network needs to remain on finding the cure while enhancing care along the way. Workshop participants expressed strong support for articulating the benefits of an IBD network to patients and their families.
What Are the Advantages of a Canadian Pediatric IBD Network?
Workshop participants discussed the advantages that Canada brings to the creation of such a network and the merits of a Canadian-only network in contrast to partnering with existing international networks. There was considerable agreement among the participants on the benefits and advantages of a Canadian network.
There are 16 pediatric centers in Canada, all of which are research oriented. These centers treat approximately 85% of pediatric patients with IBD. The relative absence of community and smaller centers providing pediatric IBD care means little care of affected children is provided outside a university hospital setting. The centralized care of pediatric patients with IBD, combined with Canada's comprehensive health care coverage, means that health researchers could access data from a representative population-based sample. Thus, it is anticipated that the majority of affected children and teenagers in Canada will be enrolled in the pediatric IBD network. Participants agreed that it is important not to exclude small centers from the network because they provide needed diversity, which is a strength of the Canadian system. Similarly, because the pediatric centers cover remote and rural areas in Canada, there is the possibility of getting unique data from these areas (including Nunavut, Yukon, and the Northwest Territories).
It was also noted that although it has a small patient population relative to other countries, Canada's comprehensive health care coverage, coupled with the high incidence of IBD, will allow researchers to collect data from diverse groups of patients who differ in terms of ethnicity, dietary habits, and therapeutic approaches and outcomes. Indeed, Canadian practice patterns vary compared with those in the United States in terms of standards of care and the availability of laboratory tests. These differences in therapeutic approaches, combined with the availability of a diverse sample of patients will enable Canadian researchers to address important and potentially unique research questions.
The small size of the pediatric IBD health care and research community and the limited number of centers should also ease the translation of scientific findings into care. On the contrary, the small size of the pediatric IBD community in Canada may also present challenges in delineating different subtypes of disease (genetic, immunological, microbiome) because it is anticipated that large numbers of patients will be needed to answer research questions about specific population subgroups. Three Canadian sites are currently part of the CCFA Pro-KIIDS Network Risk Stratification Project, helping to ensure that Canadian researchers will not be isolated from colleagues in the United States and will be in a position to share data, experiences, and best practices. Participants ultimately agreed that Canada should focus initially on a Canadian network. Future collaborations with other national and international networks are strongly encouraged.
The availability of administrative and government databases provides another potential advantage for a Canadian pediatric IBD network. Canada has strong expertise in data management with the ability to link datasets, including clinical with administrative data, biological samples, and government databases related to health care costs. Although there are a number of privacy restrictions and there may be reluctance to share administrative data across provincial jurisdictions, there is increasing consensus for linking databases to address a research agenda. Additionally, Canada has highly committed IBD research and patient/family support groups providing the capacity to mobilize patients and their families in support of a Canadian pediatric IBD network.
Objective 2: Delineating the Required Infrastructure for a National Pediatric IBD Network and Data Platform
The discussion focused on the key infrastructure needs for a data platform and a biobank, including human infrastructure to collect data and samples.
Participants reached a general consensus on the elements and underlying principles required for a comprehensive data platform. The database should have clear clinical applicability and meet the needs of researchers, clinicians, and patients. The database must be easy for health care providers to use, with procedures to ensure the quality of the data. Care should be taken to ensure that the data platform will be useful to the IBD network and to the local health care centers, as well as flexible enough to adapt to future use. Participating institutions should be guaranteed access to the database, and other incentives, such as the ability to use the system to generate prescriptions and summarize individual clinician-level data, should be considered. It is vital to include a system of graded patient consents to help patients control the use of their data and to alleviate patient concerns regarding data use. A data-coordinating center with a strong lead is essential, with some flexibility built into the system to meet the needs of local coordinating centers. A coordinating center should have access to virtual expertise including biostatisticians, perhaps patterned after SUPPORT units under the SPOR model. Ongoing data quality certification and mechanisms to ensure data comparability between centers must be built into the data platform. These databases should be centrally located with an ability to link with other databases while respecting ethical and privacy issues.
Participants agreed that data collection should be hypothesis driven with the flexibility to change data collection as new hypotheses arise. Following the development of an initial minimum dataset, the data needs may vary considerably depending on the scientific goals driving the network. The database should also have a mechanism to track patients after they leave the system (ie, patients 18 years and older). The question of reimbursement both for data collectors and study coordinators and for executive leadership for the network was raised a number of times, with the consensus that this likely is not practical in Canada. This will be a challenge for the network.
Many of the same questions and challenges noted with respect to the development of a database were seen as relevant to creating a biobank, including the needs for ongoing maintenance and quality control and hypothesis-based sample collection. Clear rules for ownership, authorship, and collaboration must be set from the beginning of the network.
The creation of a biobank requires a biorepository consisting of local storage and batching capacity, requiring consistent data entry, established sample handling, and transport practices. Participants agreed that buy-in from clinicians is essential for the success of the biobank, to ensure that sufficient patient samples are collected to make the biobank economically feasible and practical for research use. The question was raised as to whether the network could buy services from an existing biobank, rather than establishing new capacity.
Participants emphasized that because biobanking is an investment in long-term research outcomes, an investment in biobanking may not produce short-term results. One participant cautioned that although the pediatric patient participation is generally good in supporting data collection and the collection of samples, the recruitment of first-generation Canadians could prove to be problematic. Sustainability of the infrastructure is an issue that will be of critical importance to local supporting institutions.
Objective 3: Governance, Accountability, and Leadership
The need for strong leadership and accountability was articulated, with the need for early identification of a strong clinical research leader seen as an imperative. Discussions focused on a set of governance principles (Fig. 1), as a starting point to defining a governance structure.
Several governance models were discussed in terms of their components and relevance to the governance of a Canadian IBD network and data platform. These models included networks under the federal government's Networks of Centres of Excellence (NCEs) program, the CHIR Drug Safety and Effectiveness Network (DSEN), which was developed in collaboration with Health Canada to increase the available evidence on drug safety and effectiveness for regulators, policymakers, health care providers, and patients and to increase Canadian capacity to undertake high-quality postmarket research, and the Michael J. Howorth IBD Genes, Environment, and Microbiome project, a Canada-wide prospective research study funded by the CCFC that seeks to determine causes for Crohn disease by studying 5000 healthy individuals between the ages of 6 and 35, who are at a higher risk for developing the disease over time.
An NCE is generally governed by a volunteer executive committee consisting of researchers, patients, private sector representatives, and representatives from Industry Canada and the host institution. The primary role of the executive committee is to ensure that the network meets the funder's requirements in terms of outcomes, including commercialization and knowledge translation of the research, as well as developing a sustainability plan for the network once the government funds are no longer available. An NCE is led by a scientific director who is not on the executive committee. The scientific director often chairs a scientific review committee, which is responsible for the peer-review process and the excellence and relevance of the network-supported scientific research. In contrast, with DSEN the national steering committee provides strategic direction and sets priorities for research undertaken through the development of a common research agenda. CIHR appoints members to the DSEN steering committee who contribute to the objectives of DSEN in the overall interests of Canadians.
Proposed Governance Model
Participants agreed that the pediatric IBD network should be accountable to the funders, patients, researchers, and the pediatric IBD community. Leaders of the network were seen as ultimately accountable for the prudent use of funds, the quality of the research, and the relevance and timeliness of the work—in essence, for the deliverables of a pediatric IBD network. Participants stressed that deliverables need to be clearly specified and agreed to up front.
A governance model put forward for a pediatric IBD network included a network advisory board, a network management committee, and a scientific/technical oversight committee. Participating centers should have a voice in the governance of the network, and a staged approach to building the network over time would be beneficial, beginning with a steering committee that would evolve into the network advisory board. Steering committee members should be scientific leaders recruited on the basis of their expertise and track record in the focus areas of pediatric IBD, with the potential for nonpediatricians and international experts to be invited to participate to fill identified gaps. Participants stressed the importance of identifying highly respected scientists with strong project management and interpersonal skills to lead the network.
The scientific/technical oversight committee, composed of experts at arm's length, funders, and other key stakeholders, should meet twice per year to ensure accountability. The governance structure should include members who reflect the stakeholder views but that are not themselves stakeholders to manage potential conflicts of interest with respect to research agendas. Development of such an external evaluation process, involving international and national expertise, was recommended as a means to ensure that an appropriate research direction is being followed.
Funds for the Canadian Pediatric IBD Network and Data Platform are provided by CHILD and CIHR-INMD. This funding opportunity will provide 5 million Canadian dollars, enough for a single grant across a 5-year period. This amount could well increase should additional funding partners participate. Funding will be provided to the successful nominated principal investigator through a grant that is directed to the supporting institution. Oversight will be provided by the scientific/technical oversight committee, as detailed above.
Several key themes were reinforced, including the importance of this initiative for patients and their families and the sense of readiness across the research community for moving forward. The need to seek partners early in the process to ensure buy-in and potentially secure additional funding was emphasized, along with the need to align early with other relevant groups and research initiatives.
The key strengths and advantages that could be leveraged to develop a Canadian pediatric IBD network and data platform include Canada's universal health care coverage, the relatively small size of the Canadian pediatric IBD community and the limited number of pediatric centers, the unique opportunity Canada offers to study ethnically diverse communities, some of which have experience higher rates of IBD, the collaborative nature of Canadian researchers, Canada's strong expertise in data management and ability to link datasets, and the capacity of the Canadian pediatric IBD community to mobilize patients and their families in support of a network.
Although no overall consensus was reached on the scientific goals for the network, there was convergence on several themes, including the microbiome, environment and gene interactions, and the systematic study of biomarkers as predictors of effective treatment.
The 2 key infrastructure needs identified were a data platform and a biobank. The required elements and the underlying principles for this infrastructure, as well as some of the key challenges to be overcome in moving forward were enunciated, including ensuring strong relations among the centralized infrastructure and local pediatric IBD centers. Early identification of strong clinical research leadership was seen as an imperative.
1. There was consensus that the present initiative will benefit children with IBD and the gastroenterology research community.
2. The research projects at the forefront of the network should be aimed at a cure.
3. There is the potential for the creation of synergies with other entities and initiatives, both nationally and internationally.
4. Current funds will support 1 large science project plus a database. With the available funds, the projects should be developed quickly, ultimately increasing the capacity for future large-scale projects in pediatric IBD research.
5. Partners should be invited to join the initiative early to improve buy-in, to secure additional funding, and to align efforts with other relevant groups and initiatives.
At the conclusion of the workshop, the CHILD Foundation announced their commitment to the establishment of a Canadian pediatric IBD network and data platform by providing support for 5 years of funding with the potential for a 5-year renewal. The CIHR-INMD have committed to launching a call for applications in the near future, reflecting the advice received during the deliberations of the workshop.
List of workshop participants: Devendra Amre (University of Montreal), John Barnard (Ohio State University), Eric Benchimol (University of Ottawa), Jeffrey Critch (Memorial University), Sylviane Forget (McGill University), Anne Griffiths (University of Toronto), Kevan Jacobson (University of British Columbia), David Mack (University of Ottawa), Megan McMahon (CIHR), Aleixo Muise (University of Toronto), Anthony Otley (Dalhousie University), Mary Parsons (CHILD Foundation), William Paterson (Queen's University), Parminder Singh (CHILD Foundation), Rohini Soni (CCFC), Aubrey Tingle (CHILD Foundation), Bruce Vallance (University of British Columbia), Thomas Walters (University of Toronto), Eytan Wine (University of Alberta), Iwona Wrobel (University of Calgary), Nick Zamora (Telus Health Solutions).
© 2012 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,