The Canadian Neonatal NetworkTM—a novel model for knowledge translation
Shoo Lee, University of Alberta
Khalid Aziz, Memorial University of Newfoundland
Arne Ohlsson, University of Toronto
Catherine M. Cronin, University of Manitoba
Francine Lefebvre, Université de Montréal
Alexander Allen, Dalhousie University
Michael Dunn, University of Toronto
The Canadian Neonatal NetworkTM
The Canadian Neonatal NetworkTM is an award-winning research and knowledge translation (KT) initiative aimed at improving the health and quality of health care for newborn babies. Comprising researchers, clinicians, and administrators from neonatal intensive care units and universities across Canada, the Network conducts evidence-based collaborative research with an emphasis on implementation of practice and policy changes. Core activities, such as providing outcomes feedback to individual hospitals, have led to significant practice improvements. The Network has also contributed to the development of policies around the allocation of neonatal resources in British Columbia and is internationally recognized as a driver for change.
The Canadian Neonatal NetworkTM is a multidisciplinary group of researchers, clinicians, and administrators from all 30 Canadian tertiary neonatal intensive care units (NICUs) and 16 universities across Canada. We conduct evidence-based collaborative research to improve the health and quality of health care for newborn babies, with an emphasis on KT and implementation of practice and policy changes. Our program of research has resulted in many findings that have had significant impacts on patients, organizations, and health care systems. In 2004, the Canadian Neonatal NetworkTM received the Knowledge Translation Award from CIHR.
The Network's specific goals are to establish a national network of multidisciplinary Canadian researchers interested in neonatal-perinatal research, establish and maintain a national neonatal-perinatal database and provide the infrastructure to facilitate collaborative research. We are also interested in studying longitudinal outcomes and variations in medical care, because NICU care is one of the largest components of child health expenditures and exhibits large variations in mortality, morbidity, and costs. Overall, we aim to develop innovative research methods that can lead to improvements in health and the quality of health care.
The Network was founded in 1995 and its membership now stands at 50 researchers, clinicians, and administrators, and 20 trainees. Funding is provided by a partnership of hospitals (for data collection) and research granting agencies (for research projects). We have also established a Neonatal-Perinatal Interdisciplinary Capacity Enhancement (NICE) Team of multidisciplinary national experts who are funded by CIHR to provide in-depth support for research conducted by the network.
The KT initiative
We use an integrated approach that combines research methodology with experienced clinical and management expertise to identify health problems, find practical solutions, and implement and evaluate them. To do this, we maintain a standardized national database of all babies admitted to Canadian NICUs. Each year, we publish an audit report that monitors outcomes, treatment practices and health trends for newborn babies, and provides feedback to individual institutions and regional health authorities. We establish research teams to address health problems identified by the audit report and by members of the network at three levels—the patient, the organization and the health care system. We are currently conducting ten peer-review funded multi-centre studies aimed at improving neonatal outcomes.
We have published over 100 articles in scientific journals and have made over 100 presentations at national and international scientific meetings. In addition, we effect KT directly through:
- Clinical care: At each hospital, a multidisciplinary team of health professionals led by Network members monitors data feedback, and develops and implements change strategies.
- Practice guidelines: We work with professional societies (e.g. Canadian Pediatric Society) to generate and evaluate practice guidelines.
- Policy and planning: We provide health authorities and provincial governments with policy recommendations through targeted reports.
Results of the KT experience
Highlights of some of our research and KT activities include:
1. Identifying variations in NICU outcomes, practices, and best practices across Canada
We published the first comprehensive description of outcomes and practices in Canadian NICUs,1 and examined outcomes of important sub-populations at particularly high risk of long-term health complications, such as extremely preterm infants.2,3 Feedback of this information through an audit report to hospitals resulted in significant action to address identified deficiencies. One hospital introduced measures that reduced the infection rate in their NICU by more than half.
We also reported significant risk-adjusted variations in mortality and morbidity among Canadian NICUs.4 These small area variations in outcomes can be used to examine the relative effectiveness of differing practices. We found that each hospital had different strengths and weaknesses, and no hospital had uniformly superior performance. This exposed the potential flaw of copying practices from reputable hospitals, since superior performance in one area does not imply superior performance in other areas.
For one condition (intraventricular haemorrhage), we identified that variation in its incidence was attributable to differences in four key NICU practices.5 We then developed analytic methods for identifying and quantifying the attributable risks associated with outcome variations at individual hospitals.6 These developments open the way for significantly improving current quality improvement methods which rely on subjective observations of "best practices."
2. Development of instruments to compare NICU outcomes
In collaboration with U.S. researchers, we developed and patented a neonatal illness severity score (SNAP-II) and showed how risk adjustment can be used to make valid comparisons of NICU outcomes.7,8 SNAP-II has become the international standard for assessing the severity of neonatal illness and comparing hospital outcomes for audit, accreditation, and quality improvement purposes. SNAP-II has also been licensed to companies for risk assessment, setting insurance premiums, and incorporation into hospital patient monitoring equipment.
We also developed the first validated instrument9 for assessing infant transport outcomes. Using this system, one hospital in Canada identified and solved a problem affecting 40% of their transported babies, who were arriving at the destination hospital with temperatures below normal.
3. EPIC (evidence-based practice identification and change) system for quality improvement
We developed a scientific, objective method of quality improvement based on evidence in the published literature, targeted intervention of specific practices for change, and the cumulative expertise of the Network.
This evidence-based practice identification and change (EPIC) system combines quantitative analysis with qualitative methods to identify barriers to change, and uses multidisciplinary teams at each hospital to facilitate practice change. By networking hospital teams, duplication of effort is avoided, lessons learned are quickly shared, enthusiasm is maintained and KT is accomplished at the ground level.
EPIC changes the paradigm of quality improvement from a subjective exercise to an objective method that can more efficiently and effectively improve quality of care. A multi-centre CIHR-funded study is underway to evaluate the impact of EPIC on quality of care in the NICU.
4. Evaluation of clinical practice guidelines
We also use the database to evaluate practice guidelines and make recommendations for new ones. For example, we found that one routine screening guideline for preterm infants was unnecessary for babies above a certain birth weight.10 These findings led the Canadian Paediatric Society to review its national screening guidelines, with the potential to halve the number of infants routinely screened, and save hospital costs of over $1 million annually.
National networks add value because they pool expertise from a wide variety of resources, permit coordinated initiatives that avoid duplication of effort and produce results that are easily generalizable.
We also developed a formula that enables a regional hospital to identify the most cost-effective transport system for its jurisdiction, taking into account local wage costs and transport needs.11 Anecdotal communications indicate that hospitals in other countries have also started using our system for making transport system decisions.
5. Recommendations for policy and planning
Using the database, we have examined the adequacy and allocation of neonatal resources in B.C. and submitted a report to the B.C. government with recommendations for change.12 We also examined and reported on the costs of neonatal and perinatal care in B.C.13,14 These reports have since been adopted by the B.C. Ministry of Health Services and the Child Health Network for the Greater Toronto Area as a basis for assessing the adequacy of high-risk perinatal capacity and for planning regional resource allocation for the future.
Networks that include researchers, clinicians, and administrators in the planning and execution of research projects, and that execute practice and policy change at the ground level, as integral objectives of the project, are effective vehicles for KT in health care. National networks add value because they pool expertise from a wide variety of resources, permit coordinated initiatives that avoid duplication of effort and produce results that are easily generalizable. Clinical and administrative health databases have important roles in health research and can significantly reduce the cost and improve the effectiveness of health research.
We often found ourselves in the no man's land between research and quality improvement, unable to access funding from either the research granting agencies or the health care system.
A major difficulty we encountered was that our initiatives were regarded by some research peer-review committee members as quality improvement and therefore not "real research," and simultaneously regarded by some health administrators as "research." Thus, we often found ourselves in the no-man's land between research and quality improvement, unable to access funding from either the research granting agencies or the health care system.
Another major difficulty has been the lack of support for maintaining the network infrastructure or the national database, from both granting agencies and hospitals. Consequently the network infrastructure is unstable and frequently disassembled and re-assembled depending on availability of financial resources, with consequent disruption to surveillance and research. Yet the network significantly reduces the cost of individual research projects by providing common infrastructure and avoiding duplication of data collection. Funding mechanisms that bridge traditional health research and management functions, and provide stability for infrastructure support, are required.
Conclusions and implications
Funding mechanisms that bridge traditional health research and management functions, and provide stability for infrastructure support, are required.
The Canadian Neonatal Network has been described as the archetype of the KT network in Canada. It has also received international recognition as a source of benchmarking data, a driver for change, and a powerful teambuilding force. In 2003, the American Academy of Pediatrics Neonatal-Perinatal Section of District VIII voted unanimously to join the Network. Requests for assistance or collaboration have also come from countries such as India, China, Malaysia, Mexico, and Australia, and we have now established the International Neonatal Collaboration to facilitate international research and collaboration.
1 Lee, S. K., D. D McMillan, A. Ohlsson, M. Pendray, A. Synnes, R. Whyte, L. Y. Chien, and J. Sale. 2000. Variations in practice and outcomes in the Canadian NICU network: 1996-1997. Pediatrics 106:1070-79.
2 Chan, K., A. Ohlsson, A. Synnes, D. S. C. Lee, L. Y. Chien, S. K. Lee, and The Canadian Neonatal Network. 2001. Survival, morbidity, and resource use of infants of 25 weeks' gestational age or less. Am J Obstet Gynecol 185:220-26.
3 Chien, L. Y., R. Whyte, K. Aziz, P. Thiessen, D. Matthew, S. K. Lee, and The Canadian Neonatal Network. 2001. Improved outcome of preterm infants when delivered in tertiary care centers. Obstet Gynecol 98:247-52.
4 Sankaran, K., L. Y. Chien, R. Walker, M. Seshia, A. Ohlsson, and S. K. Lee. 2002. Variations in mortality rates among Canadian neonatal intensive care units. CMAJ 166:173-78.
5 Synnes, A. R., L. Y. Chien, A. Peliowski, R. Baboolal, S. K. Lee, and The Canadian NICU Network. 2001. Variations in intraventricular hemorrhage incidence rates among Canadian neonatal intensive care units. J Pediatr 138:525-31.
6 MacNab, Y., Z. Qiu, P. Gustafson, C. B. Dean, A. Ohlsson, and S. K. Lee. Forthcoming. Hierarchical Bayes analysis of multilevel health services data. Health Serv Outcome Res.
7 Richardson, D., W. O. Tarnow-Mordi, and S. K. Lee. 1999. Risk adjustment for quality improvement. Pediatrics 103 (1 Suppl E): 255-65.
8 Richardson, D. K., J. D. Corcoran, G. J. Escobar, and S. K. Lee. 2001. SNAP-II and SNAPPE-II: Simplified newborn illness severity and mortality risk scores. J Pediatr 138:92-100.
9 Lee, S. K., J. A. F. Zupancic, M. Pendray, P. Thiessen, B. Schmidt, R. Whyte, D. Shorten, S. Stewart, and The Canadian Neonatal Network. 2001. Transport risk index of physiologic stability: A practical system for assessing infant transport care. J Pediatr 139:220-26.
10 Lee, S. K., C. Normand, D. McMillan, A. Ohlsson, M. Vincer, C. Lyons, for the Canadian Neonatal Network. 2001. Evidence for changing guidelines for routine screening for retinopathy of prematurity. Arch Pediatr Adolesc Med 155:387-95.
11 Lee, S. K., J. A. Zupancic, J. Sale, M. Pendray, R. Whyte, D. Brabyn, R. Walker, and H. Whyte. 2002. Cost-effectiveness and choice of infant transport systems. Med Care 40:705-16.
12 Lee, S. K., K. Cardiff, S. Stewart, and W. MacKenzie. 2002. Report on tertiary neonatal care in British Columbia. Vancouver, BC: Centre for Healthcare Innovation and Improvement.
13 Lee, S. K., and L. Anderson. 2004. Report on tertiary neonatal costs in B.C. Vancouver, BC: Centre for Healthcare Innovation and Improvement.
14 Lee, S. K., and L. Anderson. 2004. Report on tertiary perinatal costs in B.C. Vancouver, BC: Centre for Healthcare Innovation and Improvement.
- Date modified: