Sex Differences in The Epidemiology of Cardiovascular Health Over The Life Course

Abstract

Background: Studies found that there are sex differences in cardiovascular diseases with a greater incidence among men as compared to women. The reasons for these differences are currently poorly understood. Potential reasons include differences between men and women in the prevalence of sociodemographic, lifestyle, psychosocial or biomedical risk factors for cardiovascular diseases over the life course; the magnitude or direction of association between cardiovascular risk factors and cardiovascular health; using healthcare services for the diagnosis, prevention or management of cardiovascular diseases or its risk factors. Understanding these differences might help to develop better sex-specific health prevention strategies over the life course to reduce sex differences in cardiovascular health at adulthood. Therefore, the aim of this PhD is to examine sex differences in cardiovascular health over the life course. Aims: The aims of this thesis were to (1) examine changes over time in the prevalence of cardiovascular health and cardiovascular risk factors in different age groups among men and women, (2) examine the contribution childhood socioeconomic, psychosocial, lifestyle and biomedical risk factors to sex differences in carotid artery plaques and Intima Media Thickness (IMT) in adulthood, (3) explore sex differences in the association between stroke risk factors at early adulthood with carotid artery plaques, IMT and carotid distensibility at mid-adulthood and (4) examine sex difference in the use of health services for the diagnosis and management of cardiovascular diseases and diabetes from young to mid-adulthood. Methods: This thesis contains four chapters presenting studies addressing these aims. Aim 1: Data were from participants aged 18 years and older from the Australian National Health Surveys (NHS) in 2011-12 (N=15,339), 2014-15 (N=12,297) and 2017-18 (N=13,930). Sociodemographic, health-related and biomedical factors were collected from men and women in younger, middle and older ages. Cardiovascular health score based on five key risk factors including smoking, blood pressure, physical activity, diet and Body Mass Index (BMI) were calculated with items scored 0 (poor), 1 (intermediate) or 2 (ideal). Cardiovascular health scores summed items and was categorised (low [0-4], middle [5-6], high [7-10]). Sex-specific trends in cardiovascular health over time were examined with multinomial logistic regression to estimate the Relative Risk Ratio (RRR ± 95% CI) adjusted for sociodemographic factors comparing data collected in 2014-15 and 2017-18 to 2011-12. We also used sex-specific log binomial regression to estimate the Relative Risk (RR± 95% CI) to examine the trends for the prevalence of each ideal risk factor in 2014-15 and 2017-18 compared to 2011-12. Aim 2: Children in the 1985 Australian Schools Health and Fitness Survey were followed up in the Childhood Determinants of Adult Health (CDAH) study at the age 36-49 years (2014- 19, n=1,085-1,281). Log binomial and linear regressions examined sex differences in adult carotid plaques (n=1,089) or carotid IMT (n=1,281). Childhood sociodemographic, psychosocial and biomedical risk factors that might contribute to sex differences in carotid IMT/plaques were examined using purposeful model building with additional adjustment for equivalent adult risk factors in sensitivity analyses. Aim 3: Participants in the Australian CDAH study at baseline aged 26-36 years (2004-06, CDAH-1) were followed-up at 39-49 years (2014-19, CDAH-3). Baseline risk factors included smoking, fasting glucose, insulin, systolic blood pressure (SBP) and diastolic blood pressure (DBP). Carotid artery plaques, IMT and distensibility were assessed at follow up. Log binomial and linear regression with risk factor*sex interactions predicted carotid measures. Sex-stratified models adjusting for confounders were fitted when significant interactions were identified. Aim 4: Participants were from the Australian CDAH observational cohort study at baseline aged 26-36 years. Diagnostic and therapeutic Medicare health services for cardiovascular disease and diabetes were collected in a period of approximately 10 years following CDAH￾1. We used log binomial regression analyses to examine sex differences in using health services. We then examined the contributions of other factors to these sex differences by adjusting for potential covariates in the final models based on the purposeful model building. Results: Aim 1: Men had worse overall and individual cardiovascular risk factors compared to women in each survey. There were a few changes in cardiovascular health over time for most age and sex groups. There was a lower prevalence of ideal cardiovascular health in middle-aged and older men and intermediate cardiovascular health in middle-aged women in 2017-18 compared to 2011-12. Young men and women more often had ideal diet and smoking status, yet lower prevalence for ideal BMI in 2017-18 as compared to 2011-12. Middle-aged and older men less often had ideal blood pressure in 2017-18 as compared to 2011-12. Aim 2: Among 1,281 and 1,081 participants who had complete data of carotid IMT and carotid plaques respectively, women less often had carotid plaques (10%) than men (17%). The sex difference in the prevalence of plaques was reduced by adjustment for childhood school achievement and SBP. Additional adjustments for adult education and SBP further reduced sex difference. Women (mean ± SD 0.61 ± 0.07) had thinner carotid IMT than men (mean ± SD 0.66 ± 0.09). The sex difference in carotid IMT reduced with adjustment for childhood waist circumference (WC) and SBP and further reduced with adjustment for adult WC and SBP. Aim 3: Among 779 participants with complete data (50% women), there were significant interactions of smoking, SBP and glucose with sex in the associations with carotid measures at follow-up, with associations being significant in women only. Current smoking was associated with the incidence of plaques, which reduced when adjusted for sociodemographic factors, depression, and diet. Greater SBP was associated with lower distensibility adjusted for sociodemographics. Greater glucose was associated with lower distensibility, which decreased when adjusted for sociodemographics, blood pressure, depression and polycystic ovary syndrome. Aim 4: Among those who provided consent to access universal health insurance claims data (men, n=1,222 and women, n=1,636), women more often used health services (42% vs. 35%, RR 1.20 95% CI 1.09, 1.32), especially diagnostics (38% vs. 34%, RR 1.11 95% CI 1.00, 1.24) than men. Exploration of the demographic, social, economic, psychosocial and health￾related factors that may explain the higher use in women compared to men, suggested that diet, drinking alcohol and having a family history of cardiovascular disease decreased these sex differences. Conclusion: Sex differences in cardiovascular health are evident over the life course. Women have better cardiovascular health in terms of cardiovascular risk factors and structure/function (e.g. carotid plaques, IMT and distensibility) as compared to men. Various sociodemographic, behavioural and lifestyle and biomedical factors over the life course partially explained sex differences in cardiovascular health. This included that women use more health services related to the identification and management of cardiovascular diseases than men, which was partially explained by their healthier lifestyles as compared to men. Somewhat contrary to these findings was that there were some risk factors at early adulthood that were less prevalent yet were more strongly associated with cardiovascular health in women than men. The thesis has demonstrated the complexities of sex differences in cardiovascular health over the life course. The findings suggest the potential for actions targeting sex differences in cardiovascular health, particularly risk factors, at all stages of the life course. The potential for sex-specific strategies targeting health behaviour, health literacy and engagement with preventative health services from early to mid-adulthood onwards should be tested to see if they enhance the primary prevention of cardiovascular diseases in both men and women.