Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Master's Award: Canada Graduate Scholarships

University of Manitoba

$17,500.00

Dynamic Classification Models for Administrative Health Data Case Definitions: Application to Inflammatory Bowel Disease

FY 2017-18

Summary
Administrative health databases capture electronic information about contacts with the healthcare system. These databases are widely used for population-based chronic disease research and surveillance in Canada because they contain standardized diagnosis and treatment information, capture the entire population, and are maintained over long periods of time. Case definitions have been developed to identify individuals with chronic conditions in administrative data. However, due to the presence of misclassification bias in diagnosis codes, it is often challenging to construct case definitions with high accuracy. The objective of my research is to develop methods to improve the accuracy of chronic disease case definitions in administrative data by using longitudinal information on diagnoses and treatment. I will develop two dynamic classification models that classify individuals into disease, non-disease, and indeterminate categories and then update this status as new longitudinal data becomes available. I will apply these models to inflammatory bowel disease (IBD) in an illustrative example. Administrative data from Manitoba will be used to conduct the study. IBD is highly prevalent in Manitoba compared to other Canadian jurisdictions; accurate diagnosis requires multiple contacts with the health care system. My research will contribute to improved methods to identify IBD cases in administrative data, leading to more accurate population-based estimates of disease incidence and prevalence. My research will benefit national surveillance initiatives that monitor chronic health conditions over time.
Program
Master's Award: Canada Graduate Scholarships
Location
Manitoba, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database

As published

Id
386547
Name
Feely Allison Y
Region
Manitoba
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Dynamic Classification Models for Administrative Health Data Case Definitions: Application to Inflammatory Bowel Disease
Programtype2
Studentship Programs
Orgnameinp2
University of Manitoba
Cihramount2
$17,500
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Administrative Health Data; Case Definitions; Chronic Disease; Dynamic Classification; Inflammatory Bowel Disease; Longitudinal Data; Longitudinal Discriminant Analysis; Recurrent Neural Network; Surveillance
Abstract
Administrative health databases capture electronic information about contacts with the healthcare system. These databases are widely used for population-based chronic disease research and surveillance in Canada because they contain standardized diagnosis and treatment information, capture the entire population, and are maintained over long periods of time. Case definitions have been developed to identify individuals with chronic conditions in administrative data. However, due to the presence of misclassification bias in diagnosis codes, it is often challenging to construct case definitions with high accuracy. The objective of my research is to develop methods to improve the accuracy of chronic disease case definitions in administrative data by using longitudinal information on diagnoses and treatment. I will develop two dynamic classification models that classify individuals into disease, non-disease, and indeterminate categories and then update this status as new longitudinal data becomes available. I will apply these models to inflammatory bowel disease (IBD) in an illustrative example. Administrative data from Manitoba will be used to conduct the study. IBD is highly prevalent in Manitoba compared to other Canadian jurisdictions; accurate diagnosis requires multiple contacts with the health care system. My research will contribute to improved methods to identify IBD cases in administrative data, leading to more accurate population-based estimates of disease incidence and prevalence. My research will benefit national surveillance initiatives that monitor chronic health conditions over time.
Competitiondate
201712
Approvedterm2
1 yr 0 mth
Pinamesdelim
Feely, Allison Y
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $17,500~ $0~