Department of Public Health Sciences Graduate Theses

Permanent URI for this collection


Recent Submissions

Now showing 1 - 5 of 244
  • Item
    Were Perceptions of Physical Activity Impact on Mental Health Associated With Actual Mental Health During the Covid-19 Pandemic?
    (2024-07-12) Yousufi, Sameer; Public Health Sciences; Stuart, Heather; Janssen, Ian
    Background: With the onset of the COVID-19 pandemic announced in March 2020, many studies reported deteriorating changes in physical, social, and mental wellbeing. Physical activity an established determinant of social and mental health, changed during the pandemic. This thesis examined the role of positive and negative perceptions of physical activity’s impact on mental health on severe mental distress during the COVID-19 pandemic. Objectives: The first objective was to determine if there was an association between the perceived impact of physical activity on mental health status and severe mental distress. The second objective was to determine whether this relationship varied over time as the COVID-19 pandemic progressed. Methods: Cross-sectional data from April 2021 to April 2022 collected by Mental Health Research Canada (MHRC) was used, involving 24,144 adult Canadian survey respondents across seven polls. The main exposure variable was the perception of physical activity impact on mental health (positive, negative, neutral), and the outcome variable of interest was mental health status corresponding to severe levels of mental distress. The exposure variable used an ordinal scale to assess perceptions. Severe levels of mental distress were measured using the PHQ-9 and the GAD-7, recoded to reflect severe symptoms of either. Logistic regression models were created to examine associations between the main study variables of interest while controlling for covariates. A stratified analysis of the final logistic regression model was conducted to examine findings by each survey poll. Results: Negative perceptions of physical activity impact on mental health were associated with higher odds of severe mental distress when compared to neutral perceptions, with effects modified by gender but not income (Men-high income: OR= 3.27, 95% CI = [2.25, 4.76], Men low income: OR= 3.10, 95% CI = [2.27, 4.22], Women-high income: OR= 1.87, 95% CI = [1.43, 2.45], Women-low income: OR= 1.88, 95% CI = [1.49, 2.38]). However, positive perceptions of physical activity impact on mental health were not associated with severe mental distress when compared to neutral perceptions for women.
  • Item
    Examining the Intersection of Gender and Race on Youths’ Physical Activity
    (2024-06-20) Livingston, Mikayla; Public Health Sciences; Janssen, Ian
    INTRODUCTION: Regular physical activity is associated with numerous health benefits, yet many youths, particularly those from disadvantaged backgrounds, do not accumulate enough physical activity. The primary objective of this study was to explore whether the interaction of gender and race had a unique association with physical activity that differed from the sum of their main effects. METHODS: 18,725 grades 6-10 students from the 2017-2018 cycle of the Canadian Health Behaviour in School-aged Children Study were included. Sociodemographic characteristics and different types of physical activity including active transportation, curriculum-based physical activities, organized sports, outdoor play, and structured exercise were self-reported. Log-binomial regression models were used to estimate the main effects of gender and race on low physical activity participation separately within grades 6-8 and grades 9-10 students. The relative excess risk due to interaction approach determined whether gender and race demonstrated a departure from the additivity of effects. RESULTS: Girls in grades 6-8 and grades 9-10 had a higher risk of low participation in all types of physical activity than boys. Racialized students in grades 6-8 were more likely to have low curriculum-based physical activity, organized sports, outdoor play, and total physical activity than white students. The effect of race was less pronounced in grades 9-10 students as racialized students only had an increased risk for low organized sports. There were no significant gender and race interactions for low participation among grades 6-8 students (p>0.2). However, in grades 9-10, there were significant relative excess risk due to interaction results for organized sports and total physical activity. CONCLUSION: Gender and race were associated with several low physical activity outcomes. Except for organized sports and total physical activity among grade 9-10 students, the combined effects of gender and race on low PA did not surpass the sum of their individual effects.
  • Item
    Reproductive risk factors and their association with melanoma: a population-based case-control study among Ontario women
    (2024-04-29) Chakravaram, Naga Satyasree; Public Health Sciences; VELEZ, MARIA P.; RICHARDSON, HARRIET
    Objective: To determine whether there is an association between selected reproductive risk factors (polycystic ovary syndrome-PCOS, endometriosis, parity, fibroids and premature ovarian insufficiency-POI) and melanoma. Methods and materials: A population-based case-control study was conducted using health care administrative databases in the province of Ontario, Canada. Women 15-49 years diagnosed with melanoma during the period of January 1, 1997 - December 31, 2011 (Cases, n = 4578) were identified from the Ontario Cancer Registry. Each case was individually matched by age (as calendar year of birth) and census subdivision to four randomly selected controls (n = 18,312). The presence of selected reproductive risk factors was ascertained retrospectively until January 1, 1992 from the Ontario Health Insurance Plan (OHIP) database using billing diagnostic codes. PCOS was identified using OHIP codes ICD9-256 (for PCOS diagnosis), or ICD9-709 (for hirsutism), and ICD9-626 (for irregular menses). Endometriosis was identified using two ambulatory encounters coded as OHIP codes ICD9-617 or a hospital admission or a surgical procedure coded as ICD9-617 or ICD10-N80. A history of fibroids was identified using OHIP code ICD9-218, and POI as the presence of an OHIP code ICD9-627 (for menopause) before age 40 years. Information on parity was obtained from the MOMBABY database, an ICES derived cohort. The association between selected reproductive risk factors and melanoma was quantified, simultaneously, by odds ratios derived using conditional logistic regression, and adjusted for income quintiles and immigration status based on stepwise selection method (aOR). Results: Mean age at melanoma diagnosis was 38 years. Relative to controls, women with melanoma had a similar proportion of prior endometriosis diagnosis (4.3% vs. 4.6%), nulliparity (55.74% vs. 53.7%), history of fibroids (4.6% vs. 4.5%), and POI (3.7% vs. 3.8%), and a higher proportion of PCOS diagnosis (21.9% vs. 35.1%). In the adjusted models, PCOS was associated with a higher odds of melanoma (aOR 1.93, 95% CI 1.79-2.08). None of the other selected reproductive risk factors were associated with melanoma. Conclusion: PCOS is a potential reproductive risk factor associated with melanoma. Further studies are needed to confirm this association and elucidate possible mechanisms.
  • Item
    Development of a Canadian Index of Playability
    Boyes, Randall; Public Health Sciences; Janssen, Ian; Pickett, William
    This thesis describes the creation and testing of a Canadian index of the “playability” of neighbourhoods, defined as the degree to which neighbourhoods encourage unsupervised outdoor active play in children between the ages of 10 and 13. It consists of three manuscripts. The first manuscript describes the creation and testing of a novel measurement method for features of the built environment that have been associated with the propensity of children to engage in outdoor active play. Google Street View images were systematically downloaded from the area surrounding children’s homes and processed with a neural network to extract features of the built environment. The effectiveness of this method was measured against ground truth information. The second manuscript describes the initial development of a prediction model for time spent in unsupervised outdoor activity using a sample of children from three neighbourhoods in the Greater Vancouver Area. Predictor selection was based on qualitative interviews with children conducted by the Playability Project team at the University of British Columbia. A Bayesian model of play was constructed using prior probabilities informed by the things that children indicated were important for their engagement in play. The third manuscript tests and extends the model developed in manuscript two in a cross-Canadian population of children identified using the Health Behaviour in School-aged Children survey. The predictive validity of the initial model is compared against similar models trained in the HBSC data. The Playability score developed across these manuscripts shows a low correlation with play behaviour when tested outside of the training sample, but suggests avenues for future research into built environment predictors of the play behaviour of children. The score could be refined with larger-scale datasets with objective measurements of play, and future research could examine of the causal impact of these predictors using interventional trials or by taking advantage of natural experiments. An accurate index of Playability could be used to inform the development of neighbourhoods and cities that encourage outdoor active play in children, improving their health and well-being.
  • Item
    Patterns of solar ultraviolet radiation exposures and melanoma risk in Canada
    Waddingham, Claudia M.; Public Health Sciences; King, Will
    Solar ultraviolet radiation (UVR) exposure is established as the primary cause of cutaneous melanoma. It has been hypothesized that moderate time in the sun and intermittent sun exposure increases the risk of melanoma, while chronic sun exposure has no association or may even decrease the risk of melanoma. The aim of this thesis is to examine the relationship between average time in the sun, sun exposure patterns, residential ambient UVR, and occupational UVR exposure and melanoma risk in Canada. A prospective cohort study, the Canadian Partnership for Tomorrow’s Health, collected self-reported questionnaire data on sun exposure, occupational history, and other melanoma risk factors for three provincial cohorts (Alberta, Ontario, and Quebec). Individuals from the cohorts were linked to cancer registries or administrative datasets to identify melanoma cases. Average time in the sun and residential ambient UVR exposures were investigated using a cohort methodology, while a case-cohort design was utilized to assess occupational solar exposure and risk of melanoma. For assessing occupational solar exposure, an individual’s longest job held was coded to a solar job exposure matrix to assign solar exposure status. Multivariable Cox proportional hazard models with age as the time scale and adjusted for potential confounders were used to estimate hazard ratios for each investigated relationship. Neither average daily time in thesun, intermittent and chronic solar exposure patterns, nor ambient UVR were found to have a statistically significant relationship with melanoma risk. Additionally, occupational solar exposure was not associated with risk of melanoma (HRadj = 0.94; 95% CI = 0.63-1.41). Although this study found typical time in the sun was only suggestive of increasing melanoma risk and occupational solar exposure was not associated with melanoma risk, the results align with existing literature. Understanding how different patterns of solar UVR exposure affect the risk of melanoma among Canadians can help inform on strategies aimed at targeting modifiable solar behaviors to reduce the burden of melanoma.