<?xml version="1.0" encoding="UTF-8"?><?xml-stylesheet type="text/xsl" href="static/style.xsl"?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-11-21T12:12:21Z</responseDate><request verb="GetRecord" identifier="oai:macsphere.mcmaster.ca:11375/20270" metadataPrefix="dim">https://prod-ms-be.lib.mcmaster.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:macsphere.mcmaster.ca:11375/20270</identifier><datestamp>2026-09-24T07:02:23Z</datestamp><setSpec>com_11375_271</setSpec><setSpec>col_11375_272</setSpec></header><metadata><dim:dim xmlns:dim="http://www.dspace.org/xmlns/dspace/dim" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:doc="http://www.lyncode.com/xoai" xsi:schemaLocation="http://www.dspace.org/xmlns/dspace/dim http://www.dspace.org/schema/dim.xsd">
   <dim:field mdschema="dc" element="contributor" qualifier="advisor">THABANE, LEHANA</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="author">LI, GUOWEI</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="department" lang="en_US">Clinical Epidemiology/Clinical Epidemiology &amp; Biostatistics</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2016-08-30T14:03:35Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="available">2016-08-30T14:03:35Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2016</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">http://hdl.handle.net/11375/20270</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://doi.org/10.71548/11521</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract" lang="en_US">Background and objectives: &#xd;
Prediction models are useful tools in clinical practise by providing predictive estimates of&#xd;
outcome probabilities to aid in decision making. As biomedical research advances, concerns&#xd;
have been raised regarding combined effectiveness (benefit) and safety (harm) outcomes in a&#xd;
prediction model, while typically different prediction models only focus on predictions of&#xd;
separate outcomes. A second issue is that, evidence also reveals poor predictive accuracy in&#xd;
different populations and settings for some prediction models, requiring model calibration or&#xd;
redevelopment. A third issue in data analyses is whether the treatment effect estimates could&#xd;
be influenced by competing risk bias. If other events preclude the outcomes of interest, these&#xd;
events would compete with the outcomes and thus fundamentally change the probability of&#xd;
the outcomes of interest. Failure to recognize the existence of competing risk or to account&#xd;
for it may result in misleading conclusions in health research. Therefore in this thesis, we&#xd;
explored three methodological issues in prediction models and data analyses by: (1)&#xd;
developing and externally validating a prediction model for patients’ individual combined&#xd;
benefit and harm outcomes (stroke with no major bleeding, major bleeding with no stroke,&#xd;
neither event, or both stroke and major bleeding) with and without warfarin therapy for atrial&#xd;
fibrillation; (2) constructing a prediction model for hospital mortality in medical-surgical&#xd;
critically ill patients; and (3) performing a competing risk analysis to assess the efficacy of&#xd;
the low molecular weight heparin dalteparin versus unfractionated heparin in venous&#xd;
thromboembolism in medical-surgical critically ill patients.&#xd;
Methods: &#xd;
Project 1: Using the Kaiser Permanente Colorado (KPCO) anticoagulation management&#xd;
cohort in the Denver-Boulder metropolitan area of Colorado in the United States to include&#xd;
patients with AF who were and were not prescribed warfarin therapy, we used a new&#xd;
approach to build a prediction model of individual combined benefit and harm outcomes&#xd;
related to warfarin therapy (stroke with no major bleeding, major bleeding with no stroke, neither event, or both stroke and major bleeding) in patients with AF. We utilized a&#xd;
polytomous logistic regression (PLR) model to identify risk factors and then construct the&#xd;
new prediction model. Model performances and validation were evaluated systematically in&#xd;
the study.&#xd;
Project 2: We used data from a multicenter randomized controlled trial named Prophylaxis for&#xd;
Thromboembolism in Critical Care Trial (PROTECT) to develop a new prediction model for&#xd;
hospital mortality in critically ill medical-surgical patients receiving heparin&#xd;
thromboprophylaxis. We first identified risk factors independent of APACHE (Acute&#xd;
Physiology and Chronic Health Evaluation) II score for hospital mortality, and then combined&#xd;
the identified risk factors and APACHE II score to build the new prediction model. Model&#xd;
performances were compared between the new prediction model and the APACHE II score.&#xd;
Project 3: We re-analyzed the data from PROTECT to perform a sensitivity analysis based on&#xd;
a competing risk analysis to investigate the efficacy of dalteparin versus unfractionated&#xd;
heparin in preventing venous thromboembolism in medical-surgical critically ill patients,&#xd;
taking all-cause death as a competing risk for venous thromboembolism. Results from the&#xd;
competing risk analysis were compared with findings from the cause-specific analysis.&#xd;
Results and Conclusions: &#xd;
Project 1: The PLR model could simultaneously predict risk of individual combined benefit&#xd;
and harm outcomes in patients with and without warfarin therapy for AF. The prediction&#xd;
model was a good fit, had acceptable discrimination and calibration, and was internally and&#xd;
externally validated. Should this approach be validated in other patient populations, it has&#xd;
potential advantages over existing risk stratification approaches.&#xd;
Project 2: The new model combining other risk factors and APACHE II score was a good fit,&#xd;
well calibrated and internally validated. However, the discriminative ability of the prediction&#xd;
model was not satisfactory. Compared with the APACHE II score alone, the new prediction&#xd;
model increased data collection, was more complex but did not substantially improve discriminative ability.&#xd;
Project 3: The competing risk analysis yielded no significant effect of dalteparin compared&#xd;
with unfractionated heparin on proximal leg deep vein thromboses, but a lower risk of&#xd;
pulmonary embolism in critically ill medical-surgical patients. Findings from the competing&#xd;
risk analysis were similar to results from the cause-specific analysis.</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degreetype" lang="en_US">Thesis</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degree" lang="en_US">Doctor of Philosophy (PhD)</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso" lang="en_US">en</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">PREDICTION MODEL</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">OBSERVATIONAL DATA</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">CLINICAL TRIAL</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">DATA ANALYSIS</dim:field>
   <dim:field mdschema="dc" element="title" lang="en_US">METHODOLOGICAL ISSUES IN PREDICTION MODELS AND DATA ANALYSES USING OBSERVATIONAL AND CLINICAL TRIAL DATA</dim:field>
   <dim:field mdschema="dc" element="type" lang="en_US">Thesis</dim:field>
   <dim:field mdschema="others" element="access-status">open.access</dim:field>
</dim:dim></metadata></record></GetRecord></OAI-PMH>