 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
| |
|
||||||||||||||||
| |
|
||||||||||||||||
|
Volume 22, No. 2 |
Status Report Clarence Clottey, Frank Mo, Barbara LeBrun, Phillip Mickelson, Jeff Niles and Glenn Robbins
Introduction In Canada, health care is primarily the constitutional responsibility of the provinces and territories. Consequently, considerable information related to the provision of health services is available at that level of government. At the present time, only limited use is being made of these data to assess the impact of chronic diseases such as diabetes on the population of Canada.1 This article describes an innovative means of using these administrative data banks to gather local and national epidemiological data on diabetes. Background In this information era and in light of a prevailing commitment to support evidence-based decision making in health, the Government of Canada is investing resources in the improvement of the national capacity to systematically collect, analyze and disseminate accurate health information. Canada does have some rudimentary data on diabetes from disparate sources, including investigations relying on self-reported diabetes in surveys, and on mortality and hospitalization data. At present, it is not possible to consistently track diabetes-associated complications and deaths in Canada. Research has shown that current morbidity and mortality records underestimate the burden of diabetes.2 Although diabetes kills and maims primarily through its complications, diabetes complications data are not linked to other relevant databases. This lack of systematic prevalence and incidence data limits our ability to plan and evaluate prevention and control programs for this disease. The availability of accurate, comprehensive data is an important requirement of good public health policy: accurate data on diabetes will assist in projections on its burden to the health care system. The National Forum on Health1 encouraged the use of existing administrative databases to further chronic diseases research in Canada. One such example is the National Diabetes Surveillance System (NDSS), which is being funded as part of the Canadian Diabetes Strategy (CDS). The CDS is a Government of Canada initiative launched in November 1999, funded for $115 million over 5 years. Of this, $10.8 million has been directed to the development of the NDSS. Governance of the National Diabetes Surveillance System The NDSS is the result of five years of collaboration between government and non-government bodies, and this is reflected in its governance. This partnership involves Health Canada, other federal agencies, including Statistics Canada, all provincial and territorial governments, national Aboriginal groups, academic researchers, and non-government health organizations (NGOs). Of the NGOs, the Canadian Diabetes Association has been an active partner in the development of the NDSS from the very beginning, and has mobilized private-sector sponsorship, most notably that of SmithKline Beecham. Other national NGOs with a significant interest in diabetes are represented through their membership in the multi-sectoral Diabetes Council of Canada (DCC), which promotes and participates in the NDSS. The NDSS Steering Committee is headed by the Chair of the DCC and represents all partners. It meets twice a year to monitor progress and to plan strategic directions. A secretariat is responsible for administrative and operational matters. Working groups address such key issues as data validation, data quality, database access and publication, review of the scientific direction and Aboriginal issues. Health Canada coordinates and funds the activities of the working groups.
|
||
|
FIGURE 1
|
|||
|
Goal of the National Diabetes Surveillance System The NDSS will develop provincial and territorial surveillance capacities to establish a national standardized database for disseminating comparative data on rates of diabetes and its complications in Canada. This will ultimately lead to a better understanding of the magnitude of the diabetes burden in Canada. Rationale for the National Diabetes Surveillance System To date, there is no national diabetes surveillance system in Canada. There have been some smaller investigations that relied on self-reported diabetes in surveys and on mortality and hospitalization data. This has, however, led to an underestimation of the burden of diabetes for various reasons, including the unreliability of self-reporting methods, bias in the small sample sizes in surveys and the inability to track diabetes-associated complications and deaths in Canada. The use of provincial and territorial administrative health databases will improve our ability to measure the incidence and prevalence of diabetes and establish an ongoing diabetes surveillance system in Canada. The National Diabetes Surveillance System Model To maximize the use of existing data sources before creating new ones, the NDSS initially focuses on using provincial and territorial administrative databases, such as health insurance registry, medical claim and hospital discharge data for standardized diabetes surveillance. No other national level health surveillance initiative has been based on these data sources to date, but the rationale for this approach in the study of diabetes should be apparent. The clinical path of diabetes, from detection to treatment and comorbidity management, makes it possible for the diabetes burden to be tracked through various client interactions with the provincial and territorial health care systems. Tracking can begin with physician visits (where diabetes is typically diagnosed and treated in the early stages), and followed, using the health insurance registry and/or vital statistics, through hospitalizations for diabetes-associated conditions (complications), procedures such as amputations and, ultimately, death. This system provides estimates of rates and rate ratio for populations with and without diabetes. Epidemiological and statistical models will be used in the comparative analysis of health services utilization for populations who have, and do not have, diabetes. Standardized methods also will be used in the determination of diabetes prevalence, incidence and mortality across provinces and territories. There are well-recognized precedents in parts of Canada for utilizing administrative data to provide diabetes-related information. Manitoba Health, for example, has published a number of articles in peer-reviewed journals that report on diabetes incidence and prevalence rates in the province using the health insurance registry, physician claims and hospital services.3 Demonstration Project The feasibility of the NDSS model and its implementation has been explored through a demonstration project undertaken in Manitoba, Saskatchewan and Alberta, in partnership with the University of Alberta. Matching funds for the project were awarded through Health Canada's Health Infostructure Support Program. This demonstration project documented some early deliverables for the NDSS in the areas of methodological innovation and standardization of diabetes surveillance across the three provinces. Challenges In exploring the use of provincial and territorial data for national diabetes surveillance, issues of data privacy and ownership are constant challenges. Implementation of the NDSS for the continuing surveillance of diabetes will involve linking already existing provincial and territorial administrative databases to develop longitudinal profiles of individual cases identified by unique health insurance numbers. For the purpose of privacy, these newly created longitudinal records will not contain names of people - only non-nominal case identifiers. Any person-identifying information remains in the province or territory where it is protected by corresponding data privacy legislation. Provincial or territorial data sent to Health Canada or any other NDSS partner will be non-nominal and will be in grouped, or aggregate, format. Next Steps As the capacity to use administrative data for diabetes surveillance becomes well established across the country over the next few years, the NDSS will likely broaden the reach of the surveillance by integrating or coordinating with other complementary sources of health information. Some of these sources could include national and community health risk-factor surveys. It is envisioned that the NDSS will need to identify some important areas for in-depth community-level investigation when existing administrative data does not provide sufficient information. Conclusion The NDSS has taken on the dual challenges of addressing the critical information gaps in our knowledge of the frequency and effects of diabetes in Canada and responding to the very real and important concerns related to an individual's right to privacy and an institution's right to control the use of data it has collected. It must also address provincial and territorial concerns about the transfer of data beyond jurisdictional boundaries. This project represents an important advance in chronic disease epidemiology and health service research in Canada. Health Canada is confident that the country-wide capacity building, the maintenance of partnerships and the improvement in the comprehensiveness of diabetes information represented by the NDSS initiative will ensure cautious but steady progress over the next few years towards an era of truly evidence-based decision-making. References
Author References Clarence Clottey, Frank Mo, Barbara LeBrun, Phillip Mickelson, Jeff Niles and Glen Robbins, Diabetes Division, Centre for Chronic Disease Prevention and Control, Public Health Agency of Canada, Health Canada. Correspondence: Dr. Frank Mo, Diabetes Division, Centre for Chronic Disease Prevention and Control, PHAC, Health Canada, Tunney's Pasture, AL: 1910C1, Ottawa, Ontario K1A 0L2; Fax: (613) 941-2633; E-mail: Frank_Mo@hc-sc.gc.ca [Previous] [Table
of Contents] [Next] |
|||
 |
|||
|
Last Updated: 2002-09-30 |  |
|
