Two major dietary patterns were identified in this population that named "Healthy" and "Unhealthy". The healthy dietary pattern was inversely associated with the risk of general and central obesity; while, the unhealthy dietary pattern was positively related.
The strengths of this study are the population-based design and high participation rate (more than 90 percent). Another strength point was adjustment of various confounders specially dieting and history of diabetes, hypertension and hyperlipidemia. Presence of these factors can be affected on usual dietary patterns; because, people inclined to changes in dietary habits and to adopt some healthier and more protective diets following advice from their doctors in order to control their weight or disease;3
furthermore, results were adjusted for the presence of cancer or CVD without changes in the estimations. In the other hand, adjustment of family history of diabetes and obesity can relatively control the effect of genetic on obesity. Finally, we adjusted the effect of demographic and socioeconomic factors. Also, physical activity was adjusted assuming that weight gain in adults depends on the balance between the expenditure and energy intake.21
Dietary patterns are highly influenced by socioeconomic factors. In our previous research,31
multivariate analysis showed that dietary patterns of Tehrani females were related to age, university degree, smoking, ethnicity, income, owning house and duration of residence in Tehran.
The dietary patterns identified in this study were similar to those found in a previous study done in Iran, using factor analysis of dietary intake. Esmaillzadeh and Azadbakht,24
identified 3 major dietary patterns in Tehrani female teachers aged 40-60 years, named "healthy" (high in fruits, vegetables, poultry, legumes, tea and whole grains), "western" (high in refined grains, red meats, butter, processed meats, high-fat dairy products, sweets and desserts, potatoes, eggs and hydrogenated fats), and "Iranian" (high in refined grains, potato, tea, whole grains, hydrogenated fats, legumes, and broth) dietary patterns, which healthy and western patterns were similar to identified dietary patterns in our study.
In the other hand, the healthy and unhealthy patterns of current study were consistent with "prudent" and "western" dietary patterns in Health Professionals' Follow-up Study,32
and "healthy" and "western" patterns in Swedish33
women. Likewise, they were comparable with "Spanish-Mediterranean" (as a healthy pattern) and "western" pattern in Spanish men and women participating in Spanish SUN project (Seguimiento Universidad de Navarra).3
However, comparison and interpretation of dietary patterns derived from different studies should be done with caution due to geographical and cultural as well as methodological variations.35
In this study, the relationship between healthy and unhealthy dietary patterns and risk of general and central adiposity were similar to findings of the study performed among 40-60 years Tehrani female teachers.24
In that study, women in the upper quintile of the healthy pattern were less likely to be generally and centrally obese, whereas, those in the upper category of western pattern had greater odds. Furthermore, a previously study in an Asian country (Japan) presented consistent results.23
In this study, the "Healthy" pattern (high in vegetables, mushrooms, seaweeds, potatoes, fish and shellfish, soy products, processed fish, fruit and salted vegetables) was related to lower risk of overweight and obesity and the "western" pattern, (high intakes of meats, fats and oils, seasonings, processed meats and eggs) were associated with an increased risk of BMI ≥ 25.
The inverse association of healthy pattern with risk of general and central obesity was similar to reported findings in American26
countries. In other studies, the inverse relation of BMI and weight gain with "whole grains, fruits and vegetables" pattern were reported in men36
participated in the Nurses' Health Study. Also, a "low-fat dairy, grains, and fruit" pattern in American women had an inverse association with annual changes in BMI and WC.10
Otherwise, the unhealthy dietary pattern was positively related to increased risk of general and central adiposity. In other studies, the western pattern (which is similar to our unhealthy pattern) showed the same results.13
Also, various dietary patterns high in "meat and pasta"30
and "refined grains and hotdogs"37
were positively related to BMI and a pattern high in "rice"18
was associated to general and central adiposity. Western content of this dietary pattern reflects the effect of nutrition transition on obesity in Iranian women.
The protective effect of healthy dietary pattern may be due to the effect of food with high fiber and complex carbohydrates, low glycemic index and low energy density such as vegetables, fruits and legumes38
and low fat consumption on appetite and food intake.40
While, the positive relation between unhealthy pattern and these conditions can be explained by over consumption of higher glycemic index carbohydrates (refined grains, sugars and sweets) which cause higher glycemic responses and increased fat synthesis and fat accumulation.21
Furthermore, these kinds of carbohydrates increase hunger and promote overeating. In the other hand, higher fat intake, which is seen in unhealthy dietary pattern, has been considered a risk factor for weight gain 41
. Some points should be considered in interpreting these findings. First, due to the cross-sectional design of the study, one cannot infer causality. Therefore, our findings need to be confirmed in future prospective studies. Second, limitations of FFQ for assessing dietary intakes should be taken into account (such as measurement errors inherent to the use of a FFQ for dietary assessment include underreporting or overreporting of general food intake, selective underreporting or overreporting of intakes of certain foods, or both).42
Third, several subjective or arbitrary decisions in the use of factor analysis need to be considered; the investigator is forced to prespecify the number of factors. Although eigen values, Scree plots, and interpretability are used to guide the investigator in determining the best factor solution, ultimately such a decision is subjective.11