Mid Sweden University

Background and objectives: Cardiovascular disease (CVD) is the single largest cause of mortality in the world. Similar to other health issues, CVD is generally affected either by individual risk factors, which may influence the risk for developing an illness or its complications, or by social indicators (social determinants of health). There is evidence from developed countries which shows that the so-called "upstream factors"—including social determinants such as political, social, spiritual, cultural, and economic factors—may affect the prevalence and incidence of CVD. Scarce evidence from studies in low- and middle-income countries also suggests that social factors may affect the distribution of CVD across population groups. However, there is a dearth of such data in Iran, where only a few small-sizedstudies have focused on the social determinants of health. Therefore, the present thesis sought to fill this gap by assessing the effects of socioeconomic status (SES) on the distribution of CVD and the relevant inequalities within the Iranian context.

Methods: This thesis is based on four studies, which used data from the Tehran Heart Center’s Databases. In Study I, a total of 44,820 patients who underwent coronary angiography at Tehran Heart Center between 2005 and 2010 were recruited. Then, their pre- and post-procedural data—including demographics, CVD risk factors, symptoms, and laboratory tests—were compared between men and women. In Study II, 6,246 patients with acutecoronary syndrome who were hospitalized between March 2004 and August 2011 were included and, based on their education and their employment status, were divided into high- and low-SES groups. Thereafter, the effect of SES on the in-hospital death of the patients was evaluated. In Study III, 20,165 patients with documented coronary artery disease who underwent coronary angiography at Tehran Heart Center were enrolled and CVD risk factors and severity (measured by the Gensini score) were assessed among the six major Iranian ethnic groups. In Study IV, 9,088 patients with acute coronary syndrome who were hospitalized at Tehran Heart Center between May 2007and June 2014 were recruited and the association between in-hospital death due to acute coronary syndrome and place of residence (rural/urban) was assessed using logistic regression adjusted for potential confounders.

Results: In this thesis, the data analyses were based on the hypothesis that there is a potential association between the different socioeconomic indicators and the selected cardiovascular outcomes. In Study I, among the recruited participant, 25,363 men and 11,995 women had coronary artery disease and the women not only were significantly older, less educated, and more overweight but also had higher blood levels of triglyceride, cholesterol, low-density lipoprotein, high-density lipoprotein, and fasting blood sugar than the men. Moreover, hypertension and diabetes mellitus showed the strongest association in the women with coronary artery disease (OR=3.45, 95% CI: 3.28to 3.61 and OR=2.37, 95% CI: 2.26 to 2.48, respectively). In addition, the frequency of post-procedural recommendations for non-invasive procedures was higher in the women than in the men (20.1% vs 18.6%; P<0.001). In StudyII, of the 6,246 recruited patients with acute coronary syndrome, 3,290individuals were considered low-SES and 2,956 high-SES individuals. In-hospital death occurred in 79 (1.26%) patients: 1.9% in the low-SES and 0.6% in the high-SES groups. After adjustment for the possible cofounders, our multivariate analysis demonstrated a significant effect of the patients’ SES on their in-hospital death and a lower in-hospital mortality rate was shown in the high-SES patients (OR=0.30, 95% CI: 0.09 to 0.98; P=0.046). In Study III, the Fars (8.7%) and Gilak (8.6%) ethnic groups had the highest frequency of having at least four simultaneous risk factors. Additionally, the mean Gensini score was lowest in the Lurs (67.5±52.8) and highest among the Gilaks (77.1±55.9). The multivariable regression analysis indicated that the Gilaks showed the worst CVD severity (β: 0.056, 95% CI: 0.009 to 0.102; P=0.018), followed by the Turks (β: 0.032, 95% CI: 0.005 to 0.059; P=0.020), and the lowest CVD severity, was detected in the Lurs (β: -0.087, 95% CI: -0.146 to -0.027;P=0.004). Study IV showed that while smoking (P=0.002), positive family history of coronary artery disease (P=0.003), higher body mass index (P=0.013),and hyperlipidemia (P=0.026) were more prevalent in the urban patients, the rural patients showed lower educational levels (P<0.001) and higher frequency of unemployment (P=0.009). Meanwhile, in-hospital death occurred in 135 (1.5%) patients: 125 (1.5%) urban and 10 (1.2%) rural. To adjust the effects of the possible confounders, we utilized the Firth regression model, which showed no significant difference regarding in-hospital death betweenthe rural and urban patients (OR=1.57, 95% CI: 0.376 to 7.450; P=0.585).

Conclusions: The aim of this thesis was to investigate the effects of social determinants (particularly SES) on CVD and its modifiable risk factors among Iranian patients. Results showed that medical treatment for CVD was more recommended (by treating physicians) to the women than the men, and the low-SES patients with acute coronary syndrome were more likely to die in the hospital than their high-SES counterparts. In addition, the thesis found heterogeneity in the distribution of the traditional risk factors for CVD as well as CVD severity in the major Iranian ethnic groups. Further, there were no differences concerning the in-hospital death rates due to acute coronary syndrome between the urban and rural patients after adjustment for the potential confounders.