Research: Pilot Projects

Pilot Projects Year 1 (2004-2005)

These projects are the stepping stones for further projects of the GENESIS network


The Genetic (Sex) Determinants of Excess Weight Gain in Men and Women and in Girls and Boys

Purpose: To identify candidate genes that are determinants of obesity in men and women.

Leaders: Dr. Pavel Hamet, Dr. Johanne Tremblay

Abstract: To test our hypothesis, we will carry out a genome-wide search for all phenotypes in both genders separately; Genome-wide scans will be performed with SOLAR in the entire set of hypertensive families for both genders separately. The scans will be carried out with a density of DNA markers <10cM. Genotyping of additional markers will increase the significance of the loci and refined their position. We will pursue our analysis by fine mapping on chromosome 1 in the 150-240 cM region with new polymorphic markers every 2 cM and will perform transmission disequilibrium test using FBAT (family based association test) program (version 1.4.1), which allows for the identification of regions with distorted allele transmission. This is consistent with expectations for a recently founded population like the one we are studying. One marker (D1S2141) provided us with quite significant (p=0.0003-0.008) allele distortion for a few obesity phenotypes, as well as insulin, renin, cholesterol and blood pressure phenotypes. Also other regions will be selected for further investigation based on the fact that they are gender specific. These regions are around 60cM on chr 17 for AVP expression. The next step will be to identify haplotype blocks, which may be different between men and women. The last step, which goes beyond the scope of the current application, is to identify their gene products through expression arrays, which were recently reported to be distinct in male and female rats.


Gender Determinants of Excess Weight Gain in Boys and Girls

Purpose: To identify behavioral and environmental determinants of obesity in boys and girls.

Leader: Dr. Jennifer O’Loughlin

Abstract: We hypothesize that there are differences in the behavioral and environmental determinants of excess weight gain in girls and boys.

We propose to investigate differences between boys and girls in the relative importance of gender (behavioral and environmental) risk factors for excess weight gain, in a prospective longitudinal study design. Subjects’ body mass index (BMI) at baseline, Time 1 (two years after baseline), and Time 2 (four years after baseline) will be computed as weight (kg)/height (m)2. Change in BMI will be computed as BMI at follow-up minus baseline BMI. Subjects will be categorized by decile of change in BMI and "excess weight gain" will be defined as a change in BMI equal to or greater than the 90th percentile change in BMI for same-age, same-sex students in the study population.

Predictors of excess weight gain will be identified in multiple logistic regression analysis conducted separately for boys and girls, in which the dependent variable is whether or not the subject was in the highest age- and sex-specific decile of change in BMI. Potential predictors to be investigated at the individual level will include sociodemographic characteristics, level of physical activity, frequency of sedentary behaviors (television viewing, video game playing), and dietary indicators. In further analyses both individual and environmental factors will be investigated concurrently using multi-level modeling techniques. "School" will be included in the final models to control for possible clustering related to homogeneity of students within schools. Analysis will be conducted first for excess weight gain at Time 1, then for excess weight gain at Time 2. Finally predictors of excess weight gain will be identified in survival analysis using data from all three time points. In subsequent analysis and based on the results of the project 1, we will integrate genetic determinants and assess the interplay between the gender and sex (genetic) determinants.


Gender and Sex Differences in Prodromal ACS Symptoms

Purpose: To identify different prodromal symptoms in men and women that may predict outcome post-acute myocardial infarction (AMI).

Leader: Dr. Colleen Norris

Abstract: We hypothesize that different prodromal symptoms for ACS predict future CHD events, health related quality of life and/or mortality in women and men.

Study subjects will be recruited through the APPROACH registry. APPROACH is a province-wide inception cohort of all adult Alberta residents undergoing cardiac catheterization for ischemic heart disease. The APPROACH project was initiated to study provincial outcomes of care and to facilitate quality assurance/quality improvement for patients with CAD in Alberta.48 Briefly, the APPROACH database contains detailed clinical information collected at catheterization and treatment on adult patients with known or suspected CAD. Patients in APPROACH are followed longitudinally after cardiac catheterization, thus allowing for assessment of subsequent procedure use as well as outcomes such as mortality and quality of life. Furthermore, administrative data from the discharge abstract of the hospitalization for the index catheterization, coded according to the International Classification of Disease (ICD) - 10th version is obtained and merged with the socio-demographic and clinical data collected at catheterization.49 A total of 2642 women had an index (first catheterization) catheterization in the province of Alberta between January 1 2002 and December 31, 2002. Of those women, 1075 had an indication for catheterization of ACS with electrocardiogram changes, making them potentially eligible subjects. An equal number of men will be randomly selected from the population corresponding to the same period. A list of eligible subjects attained through APPROACH will be compared against the patients discharge diagnosis to confirm ACS status (e.g. AMI). Eligible patients will be contacted to obtain permission to participate in the study. Trained nurse research assistants (RA) will telephone subjects between 4 and 6 months after their ACS. The RA will conduct the telephone survey using the McSweeney Acute and Prodromal Myocardial Infarction Symptom Survey (MAPMISS) 50 and a standard questionnaire for AMI symptoms. The MAPMISS is a telephone-administered instrument that lists 33 prodromal symptoms that were identified in qualitative studies (Appendix E). Women rate prodromal symptoms according to intensity, frequency and time frame. Data will be entered into the APPROACH database and a summary prodromal score will be constructed from the product of the intensity and frequency for each symptom noted on the MAPMISS. Data collection will occur over a one-year period. CHD events post AMI and health related quality of life data will be collected through the APPROACH registry by means of a self-reported questionnaire mailed to patients on or near the one-year anniversary of their initial cardiac catheterization. Notification regarding death will be either through the family by return mail or through a bi-annual merge with data from the Alberta Bureau of Vital Statistics.


Sex Differences in Men and Women Following Coronary Angiography

Purpose: To compare the outcomes of men and women with coronary artery stenosis of <50%.

Leaders: Dr. Karin Humphries, Dr. Susan Kirkland

Abstract: We hypothesize that 1) women presenting for coronary angiography are more likely than men to have less than 50% stenosis in all epicardial vessels on coronary angiograms and 2) women with this level of stenosis are at increased risk of hospitalization for cardiac causes compared with men.

This study will be a cohort study based on the BC Cardiac Registries (BCCR), a population-based registry that captures detailed clinical, procedural and demographic data on all patients undergoing coronary procedures in BC. This eliminates much of the referral bias inherent in other studies, institution-specific registries and highly selected patients recruited to clinical trials. Patients identified as having normal coronary arteries will be linked to Ministry of Health Databases (BC Hospitalization Database, BC Vital Statistics and PharmaNet) to obtain information on subsequent hospitalizations, revascularization procedures, deaths, and medication use.

The study population will comprise all BC residents, 20 years of age or older, undergoing CA between July 1, 1999 and December 31, 2002. Patients with prior AMI, percutaneous intervention (PCI) or coronary artery bypass surgery (CABG) will be excluded. The extent of coronary artery disease in women and men presenting for a diagnostic coronary angiogram will be described using a semi-quantitative scoring system based on the Duke Score, including < 50% stenosis, single- (? 50% stenosis), two- and three-vessel disease, or left main disease. Specifically we will compare the proportion of women versus men who have less than 50% stenosis in all epicardial vessels.

Utilizing linkage with the BC Hospitalization database, time to hospitalization for cardiac causes will be evaluated and compared between women and men. Hospitalization for cardiac causes will include any hospitalization where the most responsible diagnosis includes AMI, unstable angina stroke or a repeat cardiac catheterization, PCI, or CABG obtained through linkage with the BCCR. To allow for differences in follow-up time, Cox proportional hazards modeling will be used. The primary analysis will be time to first major adverse cardiac event defined as MI, stroke, PCI, CABG, or all-cause mortality. Planned revascularization procedure after diagnostic catheterization will not be considered an adverse outcome. Secondary analyses will include time to hospitalization for any cardiac disease (AMI, UA, CHF, stroke) and time to repeat CA or revascularization (PCI, CABG). All analyses will be adjusted for potential confounders such as comorbidities and age.


The Effectiveness of Medication Use in Men and Women with ACS

Purpose: To compare the use of and response to evidence-based cardiac medications is different between men and women with ACS.

Leaders: Dr. Louise Pilote

Abstract: We hypothesize that the use of and response to evidence-based cardiac medications is different between men and women with ACS.

Design The study population will be identified according to the criteria outlined in previous studies. Data on the treatment and clinical outcomes of all patients who were admitted for ACS in Quebec, Ontario, and British Columbia between January 1, 1998 and December 31, 2003 will be obtained from the government administrative databases in each province. The hospital discharge summary databases used to identify ACS patients in Quebec, Ontario and BC are respectively called the following: Maintenance et Èxploitation des Données pour l’Étude de la Clientèle Hospitalière (Med-Echo), Canadian Institute for Health Information (CIHI), and the BC Patient Hospitalization Data Base. Using encrypted provincial health insurance numbers, these databases will be linked to the provincial physician and drug claims databases, which contain information on patients’ in- and out-patient diagnostic and therapeutic procedures, as well as drug prescriptions. The drug claims databases are called la Régie de l’assurance maladie du Québec (RAMQ), Ontario Drug Benefit Plan (ODB) database, and PharmaCare (older than 65 years) and PharmaNet (all patients) in Quebec, Ontario and BC, respectively.

Descriptive Phase: In each of the three provinces, all drugs prescribed will first be classified into each of the medication classes. Men and women will be classified as having received an ACS drug prescription at discharge if they filled a prescription within 30 days of discharge. We will study prescriptions for the following medications: ACE inhibitors, ?-blockers, and lipid-lowering drugs. We will then compare the prescriptions for these medications between men and women. Patients with known contraindications for each drug will be excluded from the denominator.

Analytic Phase - Class Effect of ACS Medications: Flexible survival analytical methods will be used to compare the effectiveness of different medications within the three drug classes while adjusting for relevant baseline characteristics of individual patients. We will examine effectiveness on each of the following outcomes over the total duration of follow-up: mortality, rate of readmissions for ACS and frequency of outpatient physician and emergency room visits. In the initial analyses, either baseline or current exposure to specific drugs will be represented by binary dummy variables that are fixed-in-time or time-dependent. At the second stage of analyses, a time-dependent covariate indicating a switch to a different medication (if it occurs) will be added to account for the possibility that switching may be a marker for non-response. However, based on our ongoing analyses of ACS patients, the majority of ACS patients do not switch to different medications within the medication class prescribed (e.g. 8% and 11% of patients prescribed statins and ACE inhibitors, respectively). In addition, the class effect analyses we have completed so far have been robust to including all switchers and non-switchers, and non-switchers only. Finally, in the third type of analysis we will address the possibility that apparent differences in the adjusted effectiveness of different medications may be due to systematic differences in their dosage. In this order, we will replace binary drug indicators by continuous variables representing a percentage of the average dose used in clinical trials of a respective drug. If this is not available, an alternative standard dose will be used--represented by those recommended in the Compendium of Pharmaceuticals and Specialties (CPS).78 Additionally, the dose variable will be treated as a time-dependent variable and the database will be reconfigured to reflect time-dependent changes. By offering data on up to 4 years of follow-up, our databases will provide us with a rather unique opportunity to assess the long-term therapeutic effects of particular medications in both men and women.

Significance: In conclusion, there is large under-use and misuse of evidence-based medications in both male and female ACS patients. The extent of this misuse needs to be further elucidated, the differences between the genders need to be examined, and their impact must be evaluated at the population level. Administrative databases containing data on prescriptions for ACS medications and outcomes for all ACS patients in Quebec, Ontario, and BC provide an opportunity to address this important population health question. These three provinces account for over 75% of the Canadian population, and the databases of Quebec and BC provide data for all age groups. This large database will allow us to study the vast majority of male and female ACS patients in Canada. It will enable us to study the different types and combinations of ACS medications, drug effectiveness within each class, and their effect on clinical outcomes. Finally, based on the results of this study, we will be better positioned to assess the need for and to design a women-only clinical trial post-ACS.