A Federated Data Linkage Strategy to Support Population Health Research in Canada IJPDS (2017) Issue 1, Vol 1:366 Proceedings of the IPDLN Conference (August 2016)

Main Article Content

Tanya Flanagan http://www.partnershipagainstcancer.ca/
Céline Moore http://www.partnershipagainstcancer.ca/
Carolyn Sandoval http://www.partnershipagainstcancer.ca/
Published online: Apr 19, 2017


ABSTRACT


Objectives
Canada has established a pan-Canadian cohort with over 300,000 volunteer participants aged 35-69, to support research on cancer and chronic disease. A key feature of the cohort is that participants have consented to link their cohort data with administrative datasets. This prospective cohort, representing nearly 1 in 50 Canadians in this age range will be followed for multiple decades, building a platform that supports access to timely, high-quality, data related to cancer and other chronic diseases, which will enable researchers to answer complex system questions and achieve better health outcomes for Canadians.


Approach
A baseline “core” questionnaire was administered to participants capturing information on socio-demographics, economic characteristics, personal and familial history of diseases and lifestyle and health behaviours. A re-contact questionnaire is planned for 2016 to update baseline information and add depth to specific areas and capture changes over time. To realize the full potential for this cohort to support transformative research it is crucial to be able to link this data with other provincial datasets, such as cancer registries, hospital records and mortality data.


For the most part, health data in Canada resides under the purview of health providers and, or government custodians in each of the provinces and territories. As such, an innovative federated data linkage strategy is required to link cohort data with health administrative data in each regions, adhering to existing privacy and regulatory requirements, while providing central access for researchers.


Results
To date, 40% of all cohort participants have been linked with priority provincial administrative data. Each province has its own unique data linkage challenges, requiring unique customized solutions. By the end of 2017 we anticipate that the number of participants who will have had their data linked will increase to nearly 70%. The federated data linkage strategy and infrastructure offer an innovative approach that others can learn from; however to realize the full potential of the cohort and support transformative research partnerships and collaborations are required.


Conclusion
Efforts to organize resources and establish systems for data linkage and optimize data sharing and utilization in Canada are underway and include discussions with the provincial privacy commissioners, national and provincial and territorial data custodians and other thought leaders in the field.


Objectives

The vast amount of data produced by healthcare systems both structured and unstructured, termed `Big Data' have the potential to improve the quality of healthcare through supporting a wide range of medical and healthcare functions, including clinical decision support, disease surveillance, and population health management. As the field of big data in healthcare is rapidly expanding, methodology to understand and analyze thereby enhancing and optimizing the use of this data is needed. We present priorities determined for future work in this area.

Approach

An international collaboration of health services researchers who aim to promote the methodological development and use of coded health information to promote quality of care and quality health policy decisions known as IMECCHI -proposes areas of development and future priorities for use of big data in healthcare. Thematic areas were determined through discussion of potential projects related to the use and evaluation of both structured /codeable and unstructured health information, during a recent meeting in October 2015.

Results

Several themes were identified. The top priorities included: 1) electronic medical record data exploration and utilization; 2) developing common data models and multimodal /multi-source databases from disparate sources development; 3) data quality assessment including developing indicators, automated logic checks and international comparisons; 4) the translation of ICD-10 to ICD-11 through field-testing 5) Exploration of non-physician produced/coded data; and 6) Patient safety and quality measure development.

Conclusion

A list of expert views on critical international priorities for future methodological research relating to big data in healthcare were determined. The consortium's members welcome contacts from investigators involved in research using health data, especially in cross-jurisdictional collaborative studies.

Article Details