In the present study, the prevalence of NWO was determined in the population of SUMS employees. The highest prevalence was observed when considering NWO with the PBF cut-offs proposed by WHO. On the other hand, the lowest was that according to the highest quartile of PBF, which is not widely used.
Waist circumference and related parameters such as waist-hip ratio or waist-height ratio are not precise for diagnosing NWO (19). Therefore, in this study, the body fat cut-off points suggested by WHO were used to analyze the possible relationships. Nevertheless, various cut-offs have been used around the world (19); and further investigation is required to develop more proper cut-off points for PBF in the Iranian population.
Considerable discrepancies were found in the reports of the prevalence for NWO (19). The present study found a high prevalence of NWO among university employees when compared to other investigations. It should be noted that our study population supposedly had a relatively higher rate in sedentary occupation, which can be a possible reason for the observed prevalence of NWO.
The studies determining the prevalence of NWO in Asian countries are limited. A representative study of the Chinese population used a BMI range of 18.5–23.9 kg/m2 and cut-off points of PBF ≥ 24% for males and ≥ 33% for females to recognize NWO participants. The results estimated the prevalence of 7.46%. Our select definition of NWO, i.e. BMI of 18.5–24.9 kg/m2 and PBF ≥ 25 for men and ≥ 35 for women, is close to their criterion but yielded a prevalence of 24.19% among the whole sample, which is much higher than the corresponding number in their results (20). The higher prevalence in our sample may be related to the office job, which can be a contributor to obesity (21, 22). A cross-sectional study involving women in Malaysia, a developing country, estimated that the prevalence of NWO was 19.8% (23). The researchers considered a participant NWO if she had a BMI between 18.5 and 22.9 kg/m2 and a PBF in the highest tertile (> 28.52%). Comparing the highest tertile between our study and the above-mentioned investigation, it seems that our sample had much higher body fat. To better interpret these findings, the difference between the BMI ranges of the two studies should be highlighted. This discrepancy may also be related to the sedentary environment of the office in which the university’s employees work. A higher rate of NWO was found in India by Kapoor et al. using the cut-offs of PBF ≥ 20.6% in men and ≥ 33.4% in women (24). It should be noted that they studied individuals with a high risk of diabetes and aged between 30 to 60 years. These characteristics could explain the high prevalence.
Another study in India estimated a prevalence of 16.1% among young adults (25). Researchers defined high body fat as PBF > 17.6% and > 31.6% for males and females, respectively. Although they used more liberal criteria for NWO, the participants’ young age, i.e. 18–24 years old, and also sampling from a medical college which could indicate higher health literacy, may partly justify the lower prevalence of this health condition.
With regard to the results of the studies, selecting an appropriate borderline for body fat is very challenging. A study on the data of The Korea National Health and Nutrition Examination Survey proposed a definition of obesity as more than 26% and 36% body fat for men and women respectively (26). The authors explained that these cut-off points demonstrate obesity-related cardiovascular risk factors in Korean participants. It should be noted that the study defined normal weight as a BMI between 18.5 and 22.9 kg/m2 for Asian adults. Their estimates of appropriate body fat borderlines, are in close proximity to those proposed by Li et al. for Mongolian adults (27) and also to the WHO definition that was used in this study.
The present investigation showed no significant differences in energy or macronutrient intake among study groups, not even between NWNO and OW/O. Accordingly, a population-based study in Finland didn’t find any significant differences in the intakes of macronutrients among the corresponding groups; except the consumption of protein (% of energy) by women, which was higher in overweight than NWO or lean peers (10).
Contrary to these findings, a case-control study on male students in Iran, revealed that NWO and groups with overweight or obesity both had higher calorie intake than the normal-weight group (p < 0.01) (28). This study also showed lower fiber consumption in the NWO group compared to the normal-weight individuals.
With respect to dietary behaviors, Hadaye et al. showed significant differences between NWO and NWNO groups regarding food habits; such as mid-meal snacks, skipping breakfast, intake of fish, protein, high-fiber cereals, and also restaurant visits (25). To address the specific dietary behaviors assessed in the present study, Holmback et al. observed that higher meals frequency was associated with lower central obesity in men, but not in women (29). However, the authors note that some other confounding factors were not adjusted and the results may only suggest a tendency toward healthier meal patterns in participants with an active lifestyle, thus contributing to reduced body fatness. In addition, the distinction between the methods, central obesity determined by WC and NWO diagnosed by both BMI and PBF, should be mentioned to partly clarify the dissimilarity of results.
Salt consumption as another factor, is a crucial part of dietary behaviors. A cross-sectional study performed on NHANES data indicated that higher consumption of sodium was positively associated with both body fat and central obesity (30). Indeed, every 1 gram increment of sodium intake per day was linked to a 0.44% increase of total fat percent and a raise of central obesity risk by 24%. Similar findings were presented in other studies (31, 32). Generally, foods high in sodium are also high in fat and may exacerbate overeating behaviors which could result in weight gain (30). In contrast, the present study used only subjective assessment of salt intake, not sodium, and couldn’t find a significant relationship. Considering the subjective method of assessment, the participant’s responses in this study may be biased due to their conception of “high” and “low” amounts of salt.
Regarding fried food and grilled food consumption, the current study didn’t reveal any relationship with NWO. Nevertheless, in a cohort study of relatively young adults, central adiposity was significantly associated with frequent fried food consumption (33). Likewise, another study found a positive relationship between fried food intake and central obesity, only among participants in the “highest quintile of energy intake from fried food” (34). Convincingly, as fried foods usually absorb some oil and are palatable (35), their consumption could lead to higher energy intake and an increase in weight and body fatness.
Two probabilities to further explain the lack of relationship in our study, are the effects of ethnicity and physical activity on body weight or fat status. Ethnic differences not only affect the prevalence of the condition (36) but also may cause discrepancies in related factors (37). Physical activity was not assessed in this study. Relevantly, some previous studies had shown that higher consumption of energy was not necessarily associated with higher BMI and higher body fat, but physical activity and metabolic rate were inversely associated (38). Although many other studies had agreed that higher calorie consumption is associated with obesity (39, 40), additional investigation is required to enlighten these ambiguities.
With regard to limitations, it should be noted that low physical activity is a very probable contributing factor to high body fat, but in this study we didn’t have access to the data of the physical activity. Also, the results from this sample cannot be generalized to the general population, due to office work, health literacy and other life aspects of employees in comparison to whole city population.
To the best of our knowledge, this is the first study to report NWO prevalence in Iranian employee’s population. A strength of this study is the estimation of NWO prevalence by various criteria, which helps future researchers compare the results and adopt the most favorable one for their investigations.