In this study, BSEM was used to identify both direct and indirect pathways leading to total CVD, ACS, stroke, and fatal CVDs. Such an increased or decreased correlation between the traditional risk factors was related to the influence of the confounding variables while were adjusted by indirect pathways. Although previous studies considered limited risk factors, mostly in binary scale, this study showed preventive pathways with a large number of risk factors in both continuous and binary measurement scales. The Bayesian framework was applied for SEM modelling, which is reliable, especially in small sample size studies. Moreover, the resampling method used in MCMC was used in this study (36).
Many studies were performed in the literature on CVDs. Although the identification of CVD risk factors seems to be relatively comprehensive, the precise and indirect mechanisms of the associated factors underlying CVD remains unclear (37). It is a critical gap since healthy behaviours lead to both reduced risk and more effective treatments (17,38). This is the first study in the field of CVD, to the best of our knowledge, that extensively examined the direct and indirect effects of comprehensive cardiovascular-related factors using BSEM in a population-based cohort platform.
Although the indirect effects of family history of CVD, age, lipid profiles, anthropometric indices, healthy lifestyle, quality of life, high-risk behaviours, having hypertension and high blood sugar on CVDs were not covered in the literature, many studies discussed the direct effect of such factors in which these results are in line with (9,11,12,15,38–50). Causal networks evaluate causal relationships among variables beyond partial correlations and thus play a fundamental step in risk prediction. For example, although many studies listed blood pressure as the most critical risk factor for CVD (51–55), this study did not observe such a relationship. According to the results, CVD is mostly influenced by lipid profiles, which is in line with the CVD prevention study in Japan (56) and Brisighella heart study (57).
The proposed model indicated that the quality of life and healthy lifestyle components have the highest impact on lipid profile to prevent CVD. Therefore, individuals need to improve their quality of life indicators. Taking into account the physical health, as the first step, is essential. Indeed, in the absence of physical health, other indicators of quality of life do not improve. Environmental health was the second effective factor in these results. The social relationships with other people and also mental health improvement can play an essential role in CVD prevention (58,59). This study also highlighted the health benefits of high physical activity and having a healthy diet. Healthy lifestyle components also have a strong relationship with the quality of life variable. Therefore, considering the modifiable components of a healthy lifestyle, the healthy diet and the proper amount of physical activity (at least 150 minutes of physical activity per week (60)), are necessary to prevent CVD in adults (60,61).
A healthy diet was promoted in the literature to prevent CVD (62). The Mediterranean diet is one of the diet patterns in recent years, focusing on more consumption of plant foods (e.g., seeds, nuts, beans, fruits, and vegetables) and low to moderate amounts of animal food (e.g., red meat). In this diet, chicken, fish, vegetables, fruits, olive oil, NHVO, and fibres are mostly focused. Unhealthy diets, on the other hand, including western, animal fat, and fast food diets are characterised by various food items with high trans fatty acids (TFA), saturated fatty acid (SFA), and less monounsaturated fatty acid (MUFA) and Polyunsaturated fatty acid (PUFA). Such unhealthy diets also included rice, potato, and sweets, which contain high refined carbohydrates increase TG, and reduce HDL-C, as two essential components in the Iranian lipid profile (14,62–64).
Obesity and overweight are general risk factors of CVD (48). Due to the close relationship between lipid profiles and anthropometric indices, it is necessary to take into account the improved lifestyle and quality of life to reduce the incidence of CVD. Moreover, risky behaviour endangers health and quality of life and has a positive direct effect on anthropometric indices and lipid profile. Stress, on the other hand, can have immediate cardiovascular consequences, such as heart attack and sudden cardiac arrest, which can occur even in healthy people (65). The possible role of smoking-related disorders in CVD should not be ignored either. Since stress and smoking are also modifiable factors, people can easily manage these high-risk behaviours to prevent CVDs (55,66).
Non-communicable Disease Prevention organisations consider the unhealthy diet, physical inactivity, and alcohol use as common causes for stroke, heart disease, diabetes mellitus, cancer, and lung disease (67). Also, GBD findings suggest that two-thirds of strokes occur among people under the age of 70, so they should no longer be considered as old-age diseases. It has also been reported that stroke is increasing in young and middle-aged adults, possibly due to increased metabolic risk factors such as obesity and diabetes mellitus. The main risk factors for stroke were smoking and high blood pressure. Other risk factors included obesity, high blood cholesterol, and diabetes mellitus (68). In the proposed model, anthropometric indices, which were more critical than lipid profiles and risky behaviour, play a significant role in stroke.
Researches showed that the following risk factors increase the likelihood of ACS occurrence: old age, diabetes, hypertension, tobacco use, risky behaviour, male gender, and a family history of heart disease (69). The results, on the other hand, highlighted the impact of risky behaviour that cannot be ignored. Rücker et al. stated that the main risk factors of fatal CVDs were total cholesterol, smoking, and also having higher systolic blood pressure (70). Current results showed that the lipid profile was the most critical factor for (fatal) CVDs, emphasising the control of lipid profile.
In this study, a reliable statistical approach was used on a representative population of the Eastern and Mediterranean region (EMR), aiming to guide healthcare policymakers and practitioners on identifying risk factors and evidence-based screening of people at high risk of developing CVD (Fig. 1, Tables 3, and 4). The lack of knowledge about preventable risk factors could hinder the action plans on reducing CVDs (71).
Study strengths and limitations
Having developed a series of BSEMs, it is possible to examine the associations between a large number of risk factors and risk for total CVD and its different categories. The findings of this study could be of great help for healthcare professionals in recognising strong predictors for applying effective strategies for CVDs prevention. However, this study ignored the possible effects of sleep disorders leading to dyslipidaemia and CVD. Also, the effect of environmental factors, such as air pollution indices, on CVDs incidence, was not taken into account. Moreover, time-varying covariates such as treatments were not considered since this study aimed to evaluate blood factors and lifestyle components.