In this longitudinal study, we found that increased energy intakes, more than 37% of energy, from phytochemical rich foods could prevent weight gain and decrease body adiposity in adults during 3 year follow-up. Higher intakes of whole grains and fruits appeared to benefit more than other phytochemical-rich foods in prevention of weight and body fat gain. In our study, we observed higher intakes of phytochemical-rich foods and dietary PI in older; it suggesting that older participants are more likely to have healthy diets characterized by the consumption of plant foods. These findings in agreement with other studies, demonstrate that diet quality improved with age .
The mean for yearly weight gain of the participants in this study, was estimated of 0.49 kg/y over the 3 years follow-up; as compared with the higher quartile category of PI, participants in the lower quartile had twice the mean weight gain (0.64 vs. 0.31 kg/y), during the 3-year follow-up. Even modest weight gains have been shown to increase the risk of cardiovascular, type 2 diabetes, hypertension and other chronic diseases ; therefore, a beneficial effect of high dietary PI and phytochemical-rich foods intake over a 3-year period on weight gain, as observed in this study, could be instrumental in to attenuating the development of chronic disease.
The body fat, rather than the amount of excess weight, is associated with cardio-metabolic risk factors and determines the health risks of obesity, type 2 diabetes mellitus, and cardiovascular disease . BAI, which estimates percentage of adiposity directly, has been recently proposed as a new and more reliable measurement enable to facilitating fat mass determination better than BMI [27, 31]. In the current study, dietary PI was inversely associated with change in body adiposity during the study follow-up; participants with higher dietary PI had 1.47% decrease in BAI as compared with the reference group.
Recently, Vincent et al.  in a cross-sectional study of 54 healthy young adults (18–30 y) reported that PI score was inversely related to BMI, WC, waist-to-hip ratio and body fat. There are no other data in relation to dietary PI and anthropometric measures. However several previous investigations have confirmed the beneficial effects of vegetarian diets and whole plant foods to be rich sources of phytochemicals effective in body weight management [12, 33, 34].
In addition to dietary PI, higher intake of whole grains in this study was inversely associated with 3-year changes in weight and BAI; higher intake of fruit was also related with lower weight gain. We also found weak inverse associations between higher intakes of vegetables with lower weight gain, higher intakes of nuts with lower weight gain and WC change, and higher intakes of fruits with lower abdominal fat gain and reduction in body fat during the 3-year follow-up.
Many cross-sectional studies reported that higher intake of whole grains is related with lower BMI, and smaller WC in adults ; confirmed by the results of longitudinal studies one unit difference in BMI was observed between the highest and lowest whole grain intakes [36, 37]. Higher intakes of whole grain foods also have been associated with lower abdominal fat as measured by dual-energy X ray absorptiometry, and subcutaneous and visceral adipose tissue volume [38, 39]. In addition low glycemic index and high fiber content of whole grains, high phytochemical content including phenolic compounds (Ferulic acid, vanillic acid, caffeic acid, syringic acid, cumaric acid, anthocyanidins, quinines, flavonols, chalones, flavones and flavanones) may contribute to the beneficial effects of whole grains . Fruits and vegetables are other rich sources of phytochemicals. A recent systematic review of 23 longitudinal and experimental studies indicated that higher consumption of fruits and vegetables was associated with a very gradual weight gain, and contributed to reduced adiposity in overweight and obese adults in clinical trials .
Our findings did not demonstrate a significant association between dietary legumes and 3-year changes in anthropometric measures, but the association between legumes and BAI positively intensified across increase quartiles of legumes intakes; a similar trend was observed for the association of soy intakes and 3-year BAI change. Despite legumes being considered as a component of a healthy diet , there is insufficient evidence to draw clear conclusions about the protective effect of legumes on weight . Soy Isoflavones, mainly genistein, daidzein and glycitein, have been found as anti obesity factors in cell culture and animal studies; but the results of human studies are inconsistent [44, 45].
Another noteworthy finding in our results was a significant trend in intensity of inverse relation between nuts intake and 3-year changes in weight and abdominal fat in adults. Nuts in addition to mono- and polyunsaturated fat are also rich sources of fiber, plant sterol, antioxidant vitamins and polyphenols may mediate multiple health benefits. Longitudinal follow-up studies, reported that frequent nut consumption was associated with a significantly reduced risk of weight gain [46, 47].
There are several mechanisms that could explain the favorable effects of phytochemical-rich diets against obesity. The inverse association of phytochemicals and anthropometric measures may be mediated through their effects on decrease of appetite, regulation of carbohydrate and lipid metabolism, and direct effects on adipocytes . A number of fruit and vegetable polyphenols such as quercetin, naringenin, rutin, hesperidin and resveratrol were shown to inhibit preadipocyte proliferation and induce apoptosis. Several other flavonoids also decrease adipogenesis and stimulate lipolysis in adipocytes . Some phytochemicals also act as natural ligands for peroxisome proliferator-activated receptors (PPARs), and regulate lipid metabolism in the liver, promote uptake, utilization, and catabolism of fatty acids by up-regulation of genes involved in fatty acid transport and peroxisomal and mitochondrial fatty acid β-oxidation .
To our knowledge, this is the first longitudinal population-based study on usual dietary PI and changes in weight, WC and BAI. Use of a validated FFQ to assess usual dietary intakes, and 3 year follow-up to evaluate the anthropometric changes of the participants may be considered as strengths of this study.
Some limitations of the current study should be considered; usual dietary intakes of participants were only assessed at baseline, while several evaluations of dietary intakes could be increased the validity of the results. Using the USDA FCT rather than a complete Iranian FCT is another limitation.