Grant: Foundation Grant

Accuracy; Administrative Health Data; Chronic Disease; Clinical Registries; Completeness; Electronic Medical Records; Missing Data; Statistical Models; Validation

<html> <body> <p> <strong>Overview:</strong> The vision of my research program is to strengthen chronic disease research and surveillance in Canada by harnessing the power of new pan-Canadian electronic health databases. This is an important time to focus on the optimal approaches to measure chronic disease in the population, given increasing rates for many chronic conditions and growing numbers of individuals with multiple conditions. New databases that are emerging as important tools for chronic disease research and surveillance include electronic medical records (EMRs), which contain a wealth of clinical and risk factor information, and long-term care resident assessment data, to describe the complex disease patterns of older adult populations.<br> <strong>Aims and Objectives:</strong> The aims of my research program are to:<br>1. Evaluate the quality of new electronic health databases for chronic disease research and surveillance,<br>2. Develop new methods to accurately identify disease cases from multiple electronic health databases,<br>3. Improve disease risk factor measurement and risk prediction.<br>The research objectives include comparing the quality of new and established databases for disease surveillance, testing longitudinal and meta-analytic models for disease case ascertainment, and constructing probabilistic measures of disease risk for prediction models.<br> <strong>Approach:</strong> Electronic health databases housed at the Manitoba Centre for Health Policy will initially be used to examine database quality and test the feasibility of new methods for case identification. Over the long term, this research will be extended to other provinces to assess external validity. Empirical and simulation studies will be used to construct and test new statistical methods to accurately identify disease cases and evaluate risk prediction models.<br> <strong>Expertise:</strong> This research program benefits from my expertise in working with health databases from multiple jurisdictions, risk prediction models, and longitudinal data analysis techniques. Program experts from clinical medicine, population and public health, statistics, and computer science will contribute their knowledge of databases, statistical models, and behavioral risk measures and prediction. Knowledge users from the Public Health Agency of Canada, provincial health ministries, and primary care practices, will benefit from integrated knowledge translation activities.<br> <strong>Training:</strong> Trainees will participate in national and international events and activities about big data and statistical models for observational data, engage with established research networks to facilitate their interactions with multidisciplinary teams, and participate in data- and methods-related workshops, seminars, and consultations through Manitoba&rsquo;s SUPPORT (Support for People- and Patient-Oriented Research and Trials) Unit.<br> <strong>Impact:</strong> In the short term, my research will generate new knowledge about (a) the unique contributions of EMR and resident assessment databases to inform chronic disease research and surveillance, (b) accurate methods to distinguish disease cases from non-cases in electronic health databases, and (c) probability-based models to derive risk factors from administrative health databases. Over the long term, my research will help to integrate new databases into the Public Health Agency of Canada&rsquo;s Canadian Chronic Disease Surveillance System to improve the national measurement of chronic disease prevalence and incidence.</p> </body> </html>