Using a large cohort of a Middle Eastern adults population, we demonstrated favorable time trends in the population levels of total cholesterol, HDL-C, non-HDL-C, TG, TG/HDL-C and TC/HDL-C among both genders during 10 years follow up. Such favorable trends in lipids levels and ratios could not be fully accounted for by the significant increase in consumption of lipid lowering drugs. In contrast, however, population levels of general adiposity and FPG have increased in both genders potentially leading to long term risk of cardiovascular disease.
Cardiovascular disease has long been known to be a multi-factorial disease. In 1948, the Framingham Heart Study embarked on an ambitious project in health research to identify the common factors that contribute to cardiovascular disease , over the years, careful monitoring of the Framingham Study population has led to the identification of the major CVD risk factors among which lipid measures have been widely investigated [28, 29]. Several initiatives have been launched in different countries to reduce the burden of CVD by reducing the level of its risk factors [30–32]. As such, the important question to be answered would be whether these findings could be translated to better risk factor levels in the population . Perceiving this need, several studies have attempted to explore the time trends in the CVD risk factors in the populations of different ethnic groups. Favorable trends in the lipid measures have been documented by large studies conducted in European as well as North American populations [28, 34, 35]. However, the effects of these favorable trends on the population burden of CVD have paralleled the increasing trends in the obesity and diabetes [36, 37]. As a consequence, the favorable trends in the CVD morbidity and mortality leveled off at beginning of the 21st century [11, 35]. Middle Eastern population has been estimated to harbor a great fraction of the world’s burden of diabetes and obesity [20, 38]. In fact, while America have been estimated to spend more than half of the global health expenditure on diabetes, less than 10% of the global health expenditure will be spent in the low and middle-income countries . Therefore, resorting to controlling other risk factors that are more amenable to treatment and prevention continues to be the best policy to stopping CVD. We demonstrated statistically significant and clinically meaningful favorable trends in the population levels of the lipid measures over the last decade, finding that did not change after multivariate adjustment and elimination of lipid lowering drugs users. The desirable trends for lipid measures in our population is compatible with those of other studies using cross-sectional surveys [9–11, 14] and prospective studies [13, 16–18] as well.
We observed an increase in the population levels of HDL-C simultaneously with decline in triglycerides levels, a finding in agreement with results of the Framingham study  and in contrast to those of other studies [9, 10, 12], showing simultaneous increases in TC, TGs and decreasing HDL-C.
In our study the prevalence of high cholesterol decline about 48% and 42% in men and women respectively, while the percentage of US adults with high total cholesterol decline by 27% between 1999 and 2010; furthermore it was reported that about 12% of female participants and 31% of male participants had low HDL-C . However, in our adult population the prevalence of low HDL-C was 52% for men and 26% for women, despite decreasing trend in low HDL-C, dyslipidemia still has a higher prevalence, compared to U.S adults.
Although changes in nutritional habits [40, 41], physical activity and endurance exercise [42, 43] are all known to be among important determinants of serum lipid levels; the decreasing trends in lipid levels in our population could hardly be explained by life style changes (i.e. physical activity), since it was shown that low physical activity is common in Iranian population [44, 45]. It has been shown, however, that over 30% of Iranian families are now consuming less hydrogenated oil than they did in the past [46, 47], that could possibly explain the favorable lipid trend in TLGS population during recent years. In line with our findings, cross-sectional National studies conducted by Ministry of Health and Medical Education among Iranian adult population in whole country, showed significant decrease in level of high total cholesterol (Etemad K., Center for Non-communicable Diseases Control, Ministry of Health and Medical Education, Tehran, Iran, unpublished observations).
Another factor that affected serum lipids is cigarette smoking. Craig et al. in a meta-analysis about effect of smoking on cardiovascular risk factors demonstrated that compared with non-smokers, cigarettes smokers had significantly higher TC, TG and lower concentrations of HDL-C . The review study in field of smoking between 1991 and 2007 in Iran showed that during these years smoking did not increased, which might justify the favorable trend in HDL-C level of our population .
Our study has both strengths and limitations. The strengths of the current study lie in its design as a long term community-based prospective study conducted on a large sample of Middle Eastern men and women, a region where data on secular trends in the lipid levels is lacking, also lipid profile components were measured rather than self-reported. Our findings need to be interpreted in light of its limitations as is inherent to any prospective study . First, survivor bias might have biased favorable trends towards overestimated values, i.e. individuals with possible unfavorable changes in their lipid levels might have died and thus been excluded from repeated measurements. Second, as any cohort study we cannot rule out healthy cohort effect i.e. the possibility of the effect of knowledge about the serum lipids might have affected the lifestyle or lipid drugs consumption in the participants, leading to the favorable trends in lipid levels. Third, we did not have any systematic data on the trends of nutritional behavior, physical activity and knowledge of the primary prevention in our population, consequently it is not possible to test the hypothesis that whether the trends observed could be attributable to changes in physical activity or nutrition status. Forth, the results obtained in the current study might not applicable to certain age groups including younger (less than 20 years) and older (over than 75 years) ones. Finally, our population was selected from middle-aged Middle East Caucasians and therefore we cannot make inferences beyond a similar group.