Association of Students’ Nutritional Behaviors with Insulin Resistance and Body Measurements

Introduction

Nutrition provides essential energy sources for growth, development, survival, and health maintenance. When significant changes occur in university students’ lives, the poor eating habits they acquire can negatively affect later stages of their lives, making it of great importance [1]. Positive and negative emotional states have significant effects on eating behavior [2]. Emotional appetite is eating behavior that occurs in response to emotional states.

The hormone insulin is produced by β cells in the Langerhans islets of the exocrine pancreas. It is a key anabolic hormone for tissue growth and glucose metabolism. Changes in insulin levels and modifications in insulin sensitivity are risk factors for obesity. Overeating increases the amount of insulin circulating in the blood and causes insulin resistance. Furthermore, metabolic abnormalities due to increased fat tissue associated with obesity are also responsible for insulin resistance [3,4].

Eating is a complex behavior formed by the interplay of homeostatic and hedonic hunger with cognitive and emotional factors. Hedonic hunger refers to uncontrolled eating behavior under the influence of cognition, reward, and emotional factors without any energy need [5,6].

The most commonly used method for diagnosing and classifying obesity is body mass index (BMI). BMI is calculated by dividing a person’s body weight in kilograms by the square of their height in meters (kg/m2). A BMI between 18.5 and 24.9 kg/m2 is considered normal. A BMI of <18.5 kg/m2 indicates underweight, a BMI of 25–29.9 kg/m2 indicates overweight, and a BMI of >30 kg/m2 indicates obesity. The severity of obesity is classified based on BMI levels: Stage 1=30–34.9 kg/m2, Stage 2=35–39.9 kg/m2, Stage 3 ≥40 kg/m2. Although BMI is frequently used to evaluate overweight and obesity, it can be misleading due to sex, age, muscle mass, and body type. Waist circumference (WC) measurement is particularly valuable for assessing visceral obesity. Waist circumference measurements are not affected by differences such as age, sex, and muscle mass. WC has a diagnostic value for people with BMI 25–35 kg/m2. There is no benefit in measuring waist circumference for individuals with a BMI >35 kg/m2. The International Diabetes Federation recommended using population-specific WC cutoff points for the definition of obesity when defining “metabolic syndrome” in 2005 [7]. According to the TURDEP II study published in 2013 by the Turkish Endocrinology and Metabolism Society Obesity, Dyslipidemia, and Hypertension Working Group, the waist circumference cutoff points indicating visceral obesity in Turkish adults were established as 100 cm for men and 90 cm for women [8]. Obesity continues to be a significant public health issue both in Turkey and worldwide, with increasing prevalence. This study examined the relationship between university students’ nutritional habits, anthropometric measurements, and insulin resistance. University students are particularly vulnerable to unhealthy eating behaviors due to lifestyle changes, stress, and increased independence in dietary choices. The habits acquired during this period can significantly influence long-term metabolic health, including the risk of obesity and insulin resistance. Recent studies have highlighted the role of emotional and hedonic hunger in disrupting normal energy regulation, contributing to overeating and altered insulin sensitivity. However, there is a lack of data integrating emotional eating behaviors with anthropometric and biochemical indicators in young populations. In particular, the combined evaluation of BMI, waist circumference, and HOMA-IR values remains limited in the literature. This gap hinders the development of comprehensive strategies to detect early metabolic risk in this age group. This study assessed the relationship between eating behaviors and insulin resistance in university students by examining their anthropometric characteristics and emotional appetite profiles. Understanding these associations may inform preventive strategies to improve nutrition and metabolic health in young adults.

Material and Methods

STUDY DESIGN AND SAMPLING:

The population for this cross-sectional study consisted of 9896 undergraduate students studying at Ordu University from 2022 to 2024. Using the EpiInfo 7.2.5.0 program, a sample size of 370 students was determined based on a 50% prevalence, a 95% confidence interval, a 5% margin of error, and a design effect. The study used a convenience sample of students who sought services at the university’s Education and Family Health Center. While this approach ensures a high participation rate, it may introduce selection bias, as the sample may not fully represent the general student population. This limitation should be taken into account when interpreting the results. No sampling was conducted; students who sought services at the university’s Education and Family Health Center and agreed to participate in the study were included. Participants included individuals aged 18–35 without known chronic diseases, normal thyroid function, and no physical or mental health problems. The participation rate in the study was 84.05% (311 individuals).

DATA COLLECTION:

Data were collected between February 2022 and February 2024. Researchers administered a questionnaire developed through a literature review, which inquired about socio-demographic characteristics and diseases, and a three-factor face-to-face nutrition questionnaire. Additionally, anthropometric measurements of participants were taken at the center where the study was conducted, and laboratory test results were recorded. All procedures for collecting and storing data adhered strictly to ethical research standards and institutional guidelines for the protection of human subjects. To ensure data privacy and confidentiality, all personal identifiers were anonymized before data analysis. Unique participant codes were utilized in place of names or any identifiable information. Data were stored in password-protected digital files accessible only to the research team. No identifiable information was shared or published. Participants were informed that their data would remain confidential and would only be used for academic research purposes. The informed consent forms clearly outlined the data security measures in place. Data were collected over a two-year period, during which no changes were made to the data collection procedures. However, possible seasonal variations in participants’ nutritional behaviors and anthropometric measurements should be taken into account as a limitation.

SOCIO-DEMOGRAPHIC CHARACTERISTICS:

The questionnaires asked about age, gender, faculty, class, place of residence, smoking status, alcohol consumption, regular physical activity, and use of dietary supplements. Individuals who exercised 3 times a week for at least 20 minutes each time were regarded as engaging in regular physical activity. The socio-demographic data form was developed by the authors based on frequently used variables in the existing literature and public health surveys. While no formal psychometric validation was performed, the content validity of the form was ensured through expert feedback and a pre-test conducted with a pilot group of 15 university students.

THREE-FACTOR EATING QUESTIONNAIRE (TFEQ-18):

The Three-Factor Eating Questionnaire-18 (TFEQ-18) was developed by Lauzon et al in 2004 [9]. A study of validity and reliability in Turkish was conducted by Kiraç et al in 2015. The questionnaire, consisting of 18 questions, assesses individuals’ eating habits. It includes 4 subscales that measure uncontrolled eating (questions 1, 7, 13, 14, and 17), emotional eating (questions 3, 6, and 10), cognitive restraint of eating (questions 2, 11, 12, 15, 16, and 18), and hunger sensitivity (questions 4, 5, 8, and 9). Each question has its own scoring, and the scale has no cutoff point [10].

ANTHROPOMETRIC MEASUREMENTS:

The anthropometric measurements included weight, height, and waist and hip circumference, which a single person at the Education and Family Health Center performed. Waist circumference was measured at the level of the umbilicus, from the sides of the subcostal region, while standing with arms at the side using a measuring tape. Hip circumference was measured using a measuring tape at the front, from the symphysis pubis, and at the back, from the most protruding gluteal region. Height was recorded using a height scale against a flat wall, barefoot, and in an anatomical position. Body weight was measured digitally, and heavy clothing and shoes were removed. After weight and height measurements, BMI was calculated as weight/height2 (kg/m2). After waist and hip measurements, the waist-to-hip ratio was calculated as waist circumference (cm)/hip circumference (cm). A waist-to-hip ratio of <1.0 in men and <0.85 in women is considered normal [11].

LABORATORY PARAMETERS:

Participants’ fasting plasma glucose, insulin, AST, and ALT values from the past 6 months were examined. HOMA-IR was calculated using the formula: HOMA-IR=fasting insulin (microU/ml) × fasting plasma glucose (mg/dl)/405. A typical value for HOMA-IR is <2.7; ≥2.7 indicates varying degrees of insulin resistance [11].

ETHICAL CONSIDERATIONS:

The Ordu University Clinical Research Ethics Committee approved the study with decision number 11, dated January 14, 2022. Before filling out the questionnaires, participants were informed about the purpose and content of the study, and their informed consent was obtained. Participants were assured that their data would remain confidential and be used solely for scientific purposes.

STATISTICAL ANALYSIS:

The data were analyzed using SPSS version 25.0 statistical software. In descriptive findings, categorical variables were described using frequency and percentage, while numeric variables were presented as mean and standard deviation. The adequacy of the normal distribution of the data was evaluated using the Kolmogorov-Smirnov test, histogram graphics, and skewness-kurtosis coefficients. The relationship between BMI, waist-to-hip ratio, and HOMA-R was calculated using the Spearman correlation coefficient. Analyses were conducted using Student T and Kruskal-Wallis tests to evaluate the relationship between the nutrition questionnaire and its sub-dimensions with independent variables. The statistical significance level was accepted as p<0.05.

Results

The study participants were characterized as follows: 86.2% were aged 18–25, 74.0% were female, 37.0% were studying in medical school, and 22.5% were in their third year. Among the participants, 51.7% lived in dormitories, 76.5% did not smoke, 84.9% did not use alcohol, 58.8% did not exercise regularly, and 81.0% did not take dietary supplements (Table 1). Additionally, 68.2% of the study group had a normal weight, 89.7% had a standard waist-to-hip ratio, and 73.2% had HOMA-IR values within normal limits (Table 2).

A moderate positive correlation was found between BMI and waist-to-hip ratio and HOMA-IR (r=0.47, p<0.001; r=0.33, p<0.001, respectively). There was a weak positive correlation between HOMA-IR and waist-to-hip ratio (r=0.21, p<0.001).

The scores obtained from the participants’ nutrition questionnaire and its sub-dimensions are shown in Table 3. The score from the nutrition questionnaire was 40.32±7.17 (Table 3).

The factors that may be related to the nutrition scale and its sub-dimensions are shown in Table 4. The total score of the nutrition questionnaire was higher in women (p=0.01), those who were overweight or obese (p<0.001), individuals with a risky waist-to-hip ratio (p=0.03), and those with a risky HOMA-IR (p=0.04).

The degree of uncontrolled eating was higher in the 26–35 age group (p=0.02), among those in the 4th year and above (p=0.03), among overweight or obese individuals (p=0.01), and those with a risky waist-to-hip ratio (p=0.002) and risky HOMA-IR (p=0.04).

The degree of emotional eating was higher in women (p<0.001), among those studying in health-related fields (p=0.004), in individuals who were overweight or obese (p=0.003), and in those with a risky HOMA-IR (p=0.03).

The degree of consciously restricting eating was higher in those who exercised regularly (p=0.001) and those with a standard waist-to-hip ratio (p=0.01).

The level of sensitivity to hunger was found to be higher among those studying in health-related fields (p=0.03), in individuals who were overweight or obese (p<0.001), and in those with a risky waist-to-hip ratio (p=0.009).

Discussion

Obesity is influenced by many factors, with eating behavior arising from metabolic needs (homeostatic appetite), emotional reasons (hedonic appetite), or a combination of both. Metabolic signals related to energy requirements can regulate the processing of cognitive and reward functions in the cortico-limbic system. They can also disrupt energy balance by bypassing cognitive and emotional brain homeostatic regulation [12]. This interaction in the food reward system results in excessive energy intake due to impairment in dopaminergic signals, leading to obesity [13]. Increased consumption habits without physiological hunger can heighten the risk of obesity as individuals progress due to weight gain. Therefore, identifying factors associated with hedonic hunger is essential for developing individual eating habits.

In a study by Erkaya investigating the relationship between emotional appetite and obesity among university students, 71.9% of the participants were women, 35.1% regularly exercised, and 8.5% used dietary supplements [14]. In another study examining the eating habits of university students, 63.5% were women, and 33.9% exercised [15]. Our study’s gender distribution, dietary supplement use, and exercise habits were consistent with the local literature. Factors such as intense class schedules, lack of motivation, and limited campus exercise spaces may contribute to students’ low exercise rates. Janiczak et al observed that the “Food Tasted” subscale of the Power of Food Scale was significantly associated with both diet quality and body image among physically active university students; however, no significant correlation was found with BMI [16]. Low exercise rates can affect their physical and mental health, potentially triggering hedonic causes of obesity. Exercise appears to influence not only energy expenditure but also sensitivity or responsiveness to food-related cues. While the importance of exercise has increased in Turkish culture in recent years, it may still not be sufficient for this age group.

A study by Aras found that 65.6% of the students were of normal weight [15]. Another study investigating hedonic hunger among students found that 67.3% of the participants were of normal weight [17]. In another study examining university students’ attitudes towards healthy eating, the BMI was 22.7 (20.5–26.0) kg/m2, with waist circumference measuring 76.0 (69.0–84.0) cm and a waist-to-height ratio of 0.4 [18]. Since there was no study in the literature that examined BMI, waist/hip circumference, and HOMA-IR values together, sufficient comparison could not be made. However, the BMI rates of university students studying in different cities and departments were found to be similar. Our study results are consistent with the literature in this regard, suggesting that university students have similar dietary habits due to sociocultural reasons.

In Erkaya’s study, it was found that as BMI increased in women, the scores of the three-factor dietary assessment also significantly increased. Lovan and colleagues conducted a study at a state university in south Florida, finding that women have lower eating awareness, intuitive eating, and higher emotional eating behaviors compared to men [19]. A study of university students in Poland observed that women’s emotional eating scores were higher than those of men [20]. In a separate study examining university students’ exercise and dietary habits, no statistically significant differences were detected in the total dietary scores or sub-dimensions of uncontrolled eating, conscious eating, and hunger sensitivity based on gender, but a statistically significant higher rate of emotional eating was found in women [21]. A study involving 860 university students in Turkey revealed that total and subscale scores on the Power of Food Scale (PFS) were significantly higher among males, and the likelihood of developing hedonic hunger was 2.76 times greater in females compared to males [22]. Similarly, research conducted among young adults in India indicated that women reported higher levels of hedonic hunger than men; however, this difference was not statistically significant [23]. In Saudi Arabia, findings showed that females scored significantly higher than males in both hedonic hunger and emotional eating [24]. Additionally, a meta-analysis assessing the relationship between hedonic hunger and BMI identified a weakly positive association, with this relationship being more pronounced among females. The study also highlighted that the strength of this effect may vary based on the measurement tools used [25]. Our study was consistent with the literature. The higher rates of emotional eating behaviors found in women may be due to biological rhythms.

A study by Taş and Kabaran found that normal-weight and obese individuals had higher scores for uncontrolled eating, cognitive restraint, and emotional eating behaviors compared to underweight individuals [26]. In a study that included students from the Faculty of Sports Sciences, scores for uncontrolled eating, cognitive restraint, and emotional eating behaviors were above average; however, no statistically significant differences were found based on gender and age [27]. Abdella and colleagues observed that emotional eating and uncontrolled eating scores were higher in people under age 25 years [28]. A study conducted in Poland showed that with increasing obesity, hedonic hunger also increased. Consistent with the literature, our study showed that uncontrolled eating behaviors increased in individuals with obesity [29]. A study involving 427 university students in Malaysia found that scores for uncontrolled eating and cognitive restraint, measured using the TFEQ-21 scale, were significantly higher among obese students [30]. Similarly, research conducted in Poland on sociocultural influences on eating behavior among university students reported a higher prevalence of hedonic hunger among obese individuals [31]. Recent neuroimaging research provides additional insight into how hedonic influences affect eating behavior. A meta-analysis conducted by Antypa and Hendrickse found that both hunger levels and BMI significantly impact neural responses to sweet stimuli, especially in brain regions associated with reward. These results indicate that individuals with a higher BMI may have increased sensitivity to rewards, which could worsen hedonic eating tendencies [32]. When combined with our behavioral data, these findings support the idea that neurobiological responses to appealing food cues may partly explain the connection between BMI and emotional eating in university students. A recent large-scale study conducted among university students in northern Europe (n=461) found that the interaction between subjective hunger and concerns about healthy eating significantly predicted the preference for energy-dense, nutrient-poor foods (Frontiers in Nutrition, 2024). These findings show that hedonic eating behavior is influenced not only by homeostatic hunger but also by cognitive and attitudinal factors related to nutrition. Specifically, students who had lower levels of concern about healthy eating showed a stronger connection between feelings of hunger and the choice of unhealthy foods [33]. A study involving 252 Spanish university students found that greater adherence to the Mediterranean diet was linked to lower scores in emotional eating and anxiety (29% vs 23.6%). This finding supports our hypothesis that diet quality affects not just energy balance but also behaviors related to hedonic and emotional eating [34]. These results suggest a causal relationship between obesity and hedonic hunger. Including this causal relationship in the treatment protocol for patients with obesity, evaluated in a biopsychosocial approach, can facilitate and enhance the effectiveness of obesity treatment. Therefore, changes could be considered regarding evaluation of obesity monitoring conducted within the scope of preventive health services, as obesity issues can be encountered in individuals of all ages.

Abdella et al found no significant differences in scores for cognitive eating restraint [28]. In İzgi’s study, similar age groups were examined, and no statistically significant differences in the degree of food restriction based on age and gender were found [27]. Biberoğlu’s study indicated that students who did not engage in physical activity had high scores for body–food-choice alignment. In contrast, those active in physical activities showed a positive increase in eating control scores [35]. Özsoy’s study of physical activity levels and eating habits of health services vocational school students found that students had low physical activity levels and moderately healthy eating habits [36]. A study conducted in Saudi Arabia found that regular physical activity has a protective effect on individuals. Those who engaged in exercise exhibited significantly lower BMI and emotional eating scores [24]. Additionally, a meta-analysis investigated the impact of exercise and other lifestyle interventions on eating behaviors. The results showed that programs combining exercise and dietary changes significantly decreased emotional eating scores and led to notable improvements in uncontrolled eating and cognitive restraint levels [25]. Similarly, a study from Malaysia indicated that individuals who regularly participated in physical activity had significantly higher cognitive restraint scores [30]. Suliga investigated the relationship between low physical activity and the components of metabolic syndrome in adults, especially among those with a higher BMI, focusing on lifestyle factors after metabolic disturbances had already developed. In contrast, our study examined a younger university population and shows that regular physical activity not only promotes physical well-being but also helps to reduce tendencies toward emotional eating, which is a behavioral risk factor for obesity. While Suliga’s findings highlight the consequences of inactivity, our results emphasize the importance of early prevention through behavioral intervention. From a public health perspective, our study underscores the need for implementing preventive strategies within primary care and campus health services before metabolic conditions develop [37]. A comprehensive cross-sectional study conducted among university students in the United Arab Emirates and Bahrain (N=565) found that increases in hedonic hunger were strongly associated with poorer sleep quality (OR=1.43 for each unit increase in the Pleasant Food Sensitivity scale) and elevated stress levels. These findings support our observation that reduced physical activity is linked to increased emotional eating. They also highlight psychological factors – such as stress and sleep disturbances – that may enhance sensitivity to food rewards [38]. Recent studies continue to highlight the complex relationship between body mass index (BMI), gender, physical activity, and emotional eating. A study conducted by Öztürk et al involving female university students in Turkey found a significant association between higher emotional eating scores, elevated BMI, lower body satisfaction, and the presence of negative emotions. Notably, students who participated in regular physical activity reported lower emotional eating scores, suggesting that physical activity helps mitigate maladaptive eating behaviors influenced by emotional states [39]. Similarly, a systematic review and meta-analysis by Vasileiou et al indicated that emotional eating is significantly more prevalent among individuals with obesity compared to those with normal or overweight status [40]. Additionally, their findings confirmed that women are more likely to engage in emotional eating than men, consistent with previous reports. This emphasizes the impact of psychosocial factors such as body surveillance, self-esteem, and pressure related to appearance on disordered eating patterns. When we consider these findings alongside our data, it becomes clear that we must design multidimensional interventions that simultaneously address psychological, behavioral, and sociocultural influences. Our study and other studies show that university students’ exercise frequency is relatively low, but students who exercise show positive eating attitudes. Considering that regular physical activity significantly affects the immune system, emotional state, and metabolism, eating attitudes are expected to be positively influenced, considering individuals’ reasons for regular exercise, such as adopting a healthy lifestyle and caring about their appearance.

Due to the use of a convenience sampling strategy, our sample may not fully represent the entire undergraduate population, introducing potential selection bias. Future studies using randomized or stratified sampling methods could yield more generalizable results.

This study sampled students from the University Family Health Center, which facilitated high participation but may have included individuals with greater health awareness and distinct eating behaviors, limiting generalizability. It is important to note that the university population from which the sample was drawn includes both domestic and international students from a variety of cultural and regional backgrounds. This multicultural composition may help reduce the homogeneity often found in convenience sampling and improve the representativeness of the findings.

A significant positive association was found between BMI and hedonic hunger, suggesting that eating behaviors are influenced by both physiological and reward-based mechanisms. However, the lack of controls for confounders such as physical activity, sleep duration, stress, and socioeconomic status means causal interpretations should be cautious.

Moderate correlations were also found between BMI, waist-to-hip ratio (WHR), and HOMA-IR scores, indicating a link between central adiposity and insulin resistance. These findings underscore the importance of these indicators for identifying metabolic risk.

In terms of public health, accessible anthropometric screenings can identify at-risk young adults, making regular BMI and waist circumference measurements in university settings beneficial. Implementing annual screening programs alongside promoting balanced meal options and campus physical activity initiatives could further support students’ metabolic health. Future research should utilize randomized or stratified sampling methods and include participants from multiple universities or institutions to enhance the external validity and generalizability of the findings.

Conclusions

LIMITATIONS OF THE STUDY:

This study has several limitations. First, convenience sampling may limit the applicability of the findings to the broader university student population, as most participants were recruited from a single Family Health Center, potentially reflecting individuals more engaged with their health. Second, the cross-sectional design cannot establish causality; observed relationships should be viewed as associations rather than direct cause-and-effect. Additionally, variables like physical activity, sleep, stress, and eating behaviors were measured through self-report questionnaires, which can introduce inaccuracies due to memory limitations or social desirability bias. The study included university students who sought services at the family health center for any reason. The HOMA-IR value for insulin resistance was calculated by examining the participants’ medical records. The absence of registered laboratory values for all participants resulted in data loss. Finally, the sample’s specific regional and cultural context may restrict the findings’ relevance to university students in other areas or educational settings. While the university environment featured a diverse mix of students from various regions of Türkiye and international backgrounds, it does not fully represent the wider cultural and institutional differences found across different universities. Therefore, future studies should focus on including a more geographically and culturally varied sample to improve the generalizability of the findings.