Grant: Master's Award: Canada Graduate Scholarships
University of Manitoba
$17,500.00
Statistical Methods for Combining Diagnostic Validity Estimates in Administrative Health Data: Application to Juvenile Rheumatic Disease
EF 2014-15
Résumé
In Canada, as in other countries around the world, integrating and linking many types of information from different sources is helping to create rich data repositories that are useful for population health research, including chronic disease research and surveillance. Administrative health databases, which capture information about patient contacts with the healthcare system, are included in these data repositories. Administrative databases are an invaluable resource because they are relatively inexpensive to use and contain data for entire populations over time. However, the quality of these databases is often less than optimal. My research aims to develop the best methods for accurately identifying diagnosed cases of chronic disease in administrative health databases. I will be using simulation techniques to compare two statistical models for testing differences in the sensitivity (i.e., the proportion of correctly identified positive disease cases) and specificity (i.e., the proportion of correctly identified negative disease cases) for chronic disease diagnostic case definitions. Then I will apply the most powerful and least biased method(s) to an existing dataset from Manitoba, to accurately identify diagnosed cases of juvenile rheumatoid arthritis (JRA). This is the most common type of arthritis in children under the age of 17; it causes joint pain, swelling and stiffness. My methodological research will contribute to accurate estimates of JRA amongst children and youth, which ultimately will lead to a better understanding of the impact of this disease on the healthcare system, and improvements in disease treatment and management.
Programme
Master's Award: Canada Graduate Scholarships
Bénéficiaire
Emplacement
Manitoba, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
331557
Name
Kroeker Kristine
Region
Manitoba
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Statistical Methods for Combining Diagnostic Validity Estimates in Administrative Health Data: Application to Juvenile Rheumatic Disease
Programtype2
Studentship Programs
Orgnameinp2
University of Manitoba
Cihramount2
$17,500
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Administrative Health Databases; Chronic Disease; Diagnoses; Juvenile Rheumatoid Arthritis; Random-Effects Models; Sensitivity; Simulation; Specificity; Validation
Abstract
In Canada, as in other countries around the world, integrating and linking many types of information from different sources is helping to create rich data repositories that are useful for population health research, including chronic disease research and surveillance. Administrative health databases, which capture information about patient contacts with the healthcare system, are included in these data repositories. Administrative databases are an invaluable resource because they are relatively inexpensive to use and contain data for entire populations over time. However, the quality of these databases is often less than optimal. My research aims to develop the best methods for accurately identifying diagnosed cases of chronic disease in administrative health databases. I will be using simulation techniques to compare two statistical models for testing differences in the sensitivity (i.e., the proportion of correctly identified positive disease cases) and specificity (i.e., the proportion of correctly identified negative disease cases) for chronic disease diagnostic case definitions. Then I will apply the most powerful and least biased method(s) to an existing dataset from Manitoba, to accurately identify diagnosed cases of juvenile rheumatoid arthritis (JRA). This is the most common type of arthritis in children under the age of 17; it causes joint pain, swelling and stiffness. My methodological research will contribute to accurate estimates of JRA amongst children and youth, which ultimately will lead to a better understanding of the impact of this disease on the healthcare system, and improvements in disease treatment and management.
Competitiondate
201412
Approvedterm2
1 yr 0 mth
Pinamesdelim
Kroeker, Kristine
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $17,500~ $0~