The global prevalence of overweight and obesity has risen steadily since 1975, affecting all regions (1). In particular, the Eastern Mediterranean Region (EMR) has undergone a significant nutrition transition in recent decades. Obesity stands out as a major health concern in EMR countries, including Iran (2). A systematic review and meta-analysis 2021, reported that the prevalence of overweight and obesity in Iranian adults was 35 and 21%, respectively (3). Notably, obesity rates are higher among Iranian women compared to men (4). Obesity is associated with metabolic disorders and non-communicable diseases (NCDs) such as diabetes, cardiovascular diseases, non-alcoholic fatty liver disease, and cancer (5). NCDs prevalence is increasing in the EMR countries, with NCDs being the major cause of 82% of deaths in Iran in 2020 (6). However, not all people with obesity develop metabolic complications (7). Some people with obesity are not experiencing dyslipidemia, hyperglycemia, and hypertension (8). This obesity phenotype is called metabolically healthy obesity (MHO). On the other hand, the metabolically unhealthy obesity (MUO) phenotype is the state of having at least two or more metabolic disorders (9). Based on the Karelis criteria, triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), hs-CRP, and homeostatic model assessment for insulin resistance (HOMA-IR) are involved in determining MHO (10, 11).

There is limited evidence on the determinant factors of MHO and MUO. The evidence shows that higher compliance with a healthier diet and higher physical activity (PA) levels are associated with a healthier metabolic state (12, 13). Furthermore, shorter sleepers are at a higher risk of developing metabolic complications (14).

The existing evidence shows that lifestyle-related factor including adiposity is associated with a higher risk of chronic diseases (15). Central body fat distribution and impaired adipose tissue function have been shown to be the predictors of obesity-related metabolic abnormalities (16). Also, a study reported that adults with lower levels of education and income are more likely to experience MUO (17).

Obesity is related to various lifestyle factors such as diet, PA, and sleep. As a result, several factors may simultaneously affect obesity. Lifestyle factors are often interconnected, with individuals typically adopting related lifestyle patterns (18, 19). As a result, lifestyle factors should be examined together to more effectively assess their impact on health (18, 19). A priori approach provides Information on a variety of lifestyle factors that can be described by a single score. This score could be used for assessing associations between healthy and unhealthy lifestyles and health outcomes (19, 20). While previous studies have explored the relationship between individual lifestyle factors and MHO and MUO, understanding how a combination of various lifestyle factors collectively influences obesity phenotypes remains limited. Holistic approaches such as a priori and posteriori methodologies offer a comprehensive perspective by examining lifestyle factors concurrently (21). Despite the importance of exploring the combined impact of different factors on MHO and MUO, there is a paucity of evidence in this area. A study on Lebanese adults with overweight and obesity in 2020 applied the factor analysis approach for the first time and assessed the association between lifestyle patterns and MHO and MUO and found a positive link between a healthy lifestyle pattern and MHO (22). Nevertheless, no prior study has employed an a priori approach to investigate the correlations between lifestyle scores and MHO and MUO. Consequently, this study is designed to explore the associations between lifestyle risk scores (LRS) and various obesity phenotypes among Iranian women with overweight and obesity. While Middle Eastern populations including Iranians have a particular type of obesity which is characterized by abdominal fat accumulation and greater waist circumference, the evidence on the lifestyle factors in relation to obesity phenotype is scant in this population. As a result, it is of significance to advance knowledge about this region (23).

The sample size was measured using the formula below: With α = 0.05, β = 0.95, and r = 0.25, n = [(Z 1 − α + Z 1 − β) × √1 − r2]/r 2 + 2). SPSS software version 26 was used to perform the data analysis. The Kolmogorov–Smirnov test was utilized to confirm the normal distribution of dependent variables. Continuous variables were presented as mean and standard deviation while categorical variables were presented as the number and percentage. The continuous variables between the two groups were compared using a one-way analysis of variance (ANOVA), while categorical variables were compared using the chi-square test. Analysis of covariance (ANCOVA) was used to adjust the analysis for potential confounders including age, energy intake, and BMI. A binary logistic regression analysis was used to evaluate the association between obesity phenotype categories (binary dependent variable) and LRS (continuous independent variable). The analysis was adjusted for age and energy intake. The 0.05 level of significance was considered as minimal statistical significance. A p-value <0.05 was considered statistically significant.

To the best of our knowledge, this study, for the first time, applied a priori approach to examine the LRS in relation to obesity phenotype. For developing the LRS, dietary intake, PA, sleep quality, WHtR, and SES were involved. The primary finding indicates that women with higher LRS are more likely to experience MUO.

This study demonstrated that women with higher LRS were at a greater risk of developing MUO. Our findings underscore the impact of lifestyle factors on obesity, particularly the role of PA. Women with higher LRS, indicative of unhealthier lifestyles, demonstrated lower levels of PA. The evidence consistently supports the positive influence of PA on metabolic health, with benefits including improved lipid profiles, insulin sensitivity, glucose uptake, and reduced blood pressure and inflammation (13, 47, 48). Furthermore, PA levels are found to be inversely associated with CRP levels (49). While no previous study assessed the LRS in relation to obesity phenotype, only one study employed an empirical approach and examined a combination of lifestyle factors in relation to obesity phenotype. (22) In accordance with our findings, a cross-sectional study by Naja et al. investigated 350 Lebanese adults and found a positive association between a healthier lifestyle pattern and MHO. This healthier lifestyle pattern was characterized by increased levels of PA (22). However, findings from studies examining the relationship between PA and obesity phenotype are inconsistent. While Farabi et al. and Lopez-Garcia et al. on 99 and 11,520 adults from America and Spain found that the MHO group had higher levels of PA than the MUO group (50, 51). A study on 3,807 adults from Irland did not find any difference in the levels of PA between MUO and MHO groups (52).

The positive association between higher LRS or unhealthier lifestyle and MUO might be due to dietary intake. Participants with lower LRS had a higher intake of whole grains, legumes and vegetables. Vegetables, whole grains and legumes have a high content of fibers, vitamin C, vitamin A, carotenoids, and phytochemicals, which are protective compounds against inflammation and oxidative stress (53). Consistent with our findings, Naja et al.2020 showed that a healthier lifestyle pattern characterized by a higher intake of fruits, vegetables, and legumes was positively linked to MHO (22). Furthermore, a study on 203 Iranian adolescents with overweight and obesity found that higher vegetable intake was negatively associated with MUO risk (54). Also a study on 137 adults with obesity from Brazil found that participants with MHO had a higher diet quality index (55).

A direct link between higher LRS and MUO may be explained by the sociodemgraphic characteristics of study participants. The majority of women with higher LRS had lower levels of education. There is limited evidence on the association between education levels and MHO and MUO. In line with our findings, two studies on 20,430 US adults and 11,520 Spanish adults found that participants with obesity and MUO had lower levels of education (17, 51). A study on 1,073 students from a Mexican university reported that higher education levels were associated with better metabolic health (56). There is abundant evidence indicating that a higher level of education is associated with improved health results and people with higher levels of education are more likely to follow the healthier behavior (56–59). Furthermore, the majority of participants with higher LRS had poor sleep quality. A study on 1,777 US adults reported that participants with sleep complications were less likely to be MHO. While the relationship between sleep and obesity phenotype is still unclear, the evidence shows that chronic lack of sleep results in weight gain through endocrine pathways including leptin and ghrelin which regulate appetite. Poor sleep quality is also associated with higher inflammation, increased sympathetic nervous system activity, insulin resistance, cardiometabolic dysfunction and obesity (60–62).

This study has several limitations that need to be taken into consideration. Firstly, the study design is cross-sectional which is susceptible to reverse causality when examining associations between the LRS and obesity phenotypes. Secondly, dietary intake was assessed through the FFQ, which introduces the potential for memory bias. Thirdly, it is important to note that the study participants were exclusively women from Tehran, making the sample non-representative of the entire Iranian population. Fourthly, this study used the Karelis criteria which does not account for blood pressure. This limitation should be taken into account while comparing the findings of this study with the studies that considered blood pressure to define obesity phenotype. Fifthly, this study used a priori approach to create LRS which has limitations. The indices are created based on the current knowledge and the evidence. As a result, this approach is limited by the existing evidence (20). Lastly, the majority of participants were MUO which could influence the findings of this study. This study has a number of strengths. To the best of our knowledge, for the first time, this study applied an a priori approach, created the LRS and examined associations between the LRS and obesity phenotypes. The analysis of this study was controlled for several confounding factors. Given the obesity is more prevalent in the women from the EMR than men, this study included Iranian women with overweight and obesity.

This study, used a novel a priori approach through the creation of LRS, showed significant associations between an unhealthier lifestyle, as indicated by a higher LRS, and an increased likelihood of MUO in Iranian women with overweight and obesity. The study highlights the intricate interplay between lifestyle elements and obesity phenotypes, emphasizing the need for comprehensive interventions addressing multiple factors to promote metabolic health. However, the study’s cross-sectional nature introduces limitations, and further research, ideally incorporating longitudinal designs and diverse populations, is warranted to validate these associations and enhance our understanding of the complex relationship between lifestyle and obesity.